Full Guideline 490503565

“The original NICE schizophrenia guideline was of remarkable
superiority in its methodological quality compared with other

national treatment guidelines throughout the world. This
updated version of the guideline is yet again of exceptional
quality, demonstrating rigour in its development, clarity in its
presentation and noticeable breadth in its coverage. Whether
dealing with drug and psychosocial treatments, patient
experience, ethnic minorities or health economics, based
on current evidence the guideline opens up new vistas on
the best treatments available for people with schizophrenia.
A landmark of schizophrenia practice guidelines.”
Professor Wolfgang Gaebel, MD, Professor of Psychiatry, Director,
Department of Psychiatry and Psychotherapy of the Heinrich-Heine-University, Düsseldorf
and Past President German Psychiatric Association (DGPPN)
This guideline on Schizophrenia, commissioned by NICE and developed by the

National Collaborating Centre for Mental Health, sets out clear, evidence- and
consensus-based recommendations for healthcare staff on how to manage and treat
schizophrenia in adults.
It is an update of the previous guidance (published 2002), which was the first
guideline that NICE ever produced and which was judged to be superior to other
SCHIZOPHRENIA
schizophrenia guidelines in an international survey.
This updated guideline provides new clinical and economic evidence about the use of
PSYCHOSIS and
psychological and psychosocial interventions and antipsychotic drugs and new
reviews of early intervention services, primary care and treatment for physical health
problems. There are also new chapters on access and engagement for minority ethnic
groups and on service user and carer experience of treatment and care for
Schizophrenia
schizophrenia.
An accompanying CD contains further information about the evidence, including:

health economics evidence tables
in adults
characteristics of and references for included and excluded studies
all meta-analytical data presented as forest plots
THE NICE GUIDELINE ON TREATMENT
detailed information about how to use and interpret forest plots.
AND MANAGEMENT –
UPDATED EDITION 2014 14

Cover photo: iStockphoto.com
Psychosis and
schizophrenia in adults
Treatment and management

This guideline should be read in conjunction with ‘Service user
experience in adult mental health’, NICE clinical guideline 136
National Clinical Guideline Number 178
National Collaborating Centre for Mental Health

Commissioned by the
National Institute for Health and Care Excellence
Update information
December 2021: Following a surveillance review we have updated recommendations
on monitoring for people taking antipsychotic medication to say that either glycosylated
haemoglobin (HbA1c) or fasting blood glucose may be used to test for diabetes.
July 2020: We linked to the NICE guideline on supporting adult carers in the
recommendations on providing a carer's assessment. We incorporated footnote text
into the recommendations to meet accessibility requirements.
August 2019: We amended the recommendation on offering a healthy eating and
activity programme to note that the advice has not changed after our review of the
2019 STEPWISE trial. Links have been updated.
March 2014: We corrected the wording of the recommendation on offering help to stop
smoking to clarify that it is the hydrocarbons in cigarette smoke that cause interactions
with other drugs, rather than nicotine.
These changes can be seen in the short version of the guideline at:
www.nice.org.uk/guidance/cg178
Psychosis and schizophrenia in adults 1

GUIDELINE DEVELOPMENT GROUP MEMBERS
Elizabeth Kuipers (Chair, Guideline Development Group)
Professor of Clinical Psychology, Institute of Psychiatry, King’s College London
Tim Kendall (Facilitator, Guideline Development Group)

Director, National Collaborating Centre for Mental Health; Medical Director and
Consultant Psychiatrist, Sheffield Health and Social Care NHS Foundation Trust
Amina Yesufu Udechuku

Systematic reviewer (from March 2012)
Eric Slade
Health economist (from January 2013)
Max Birchwood
Professor of Youth Mental Health, Division of Health and Wellbeing, Warwick
Medical School, University of Warwick and Director of Research, Youthspace
programme, Birmingham and Solihull Mental Health Foundation Trust
Alison Brabban
Consultant Clinical Psychologist, Tees, Esk & Wear Valleys NHS Foundation Trust;
Honorary Senior Clinical Lecturer, Durham University; National Advisor for Severe
Mental Illness (IAPT), Department of Health
Lucy Rebecca Burt

Research assistant (from October 2013)
Nadir Cheema
Health economist (until November 2012)
Debbie Green
Directorate Lead for Occupational Therapy and Social Inclusion, Adult Mental
Health, Oxleas NHS Foundation Trust, London
Bronwyn Harrison
Research assistant (until October 2013)
Zaffer Iqbal
Head of Psychology and Consultant Clinical Psychologist, Navigo NHS Health &
Social Care CiC
Sonia Johnson
Professor of Social and Community Psychiatry, Mental Health Sciences, University
College London; Consultant Psychiatrist, Camden and Islington Early Intervention
Service, Camden and Islington NHS Foundation Trust
Tom Lochhead

Mental Health Lead Professional for Social Work in Bath & North East Somerset
Max Marshall
Professor of Community Psychiatry, University of Manchester; Honorary
Consultant, Lancashire Care NHS Foundation Trust; Medical Director Lancashire
Care NHS Foundation Trust; Deputy Director/Associate Director Mental Health
Research Network England
Evan Mayo-Wilson
Senior systematic reviewer (until March 2012)
Jonathan Mitchell
Consultant Psychiatrist, Sheffield Health and Social Care NHS Foundation Trust
Tony Morrison
Professor of Clinical Psychology, Division of Psychology, University of Manchester
Maryla Moulin
Project manager
David Shiers
GP Advisor to the National Audit of Schizophrenia (the Royal College of
Psychiatrists), London; Rethink Mental Illness Trustee (2010-2012)
Sarah Stockton
Senior information scientist
Clare Taylor
Senior editor
Clive Travis
Service User Representative
Rachel Waddingham
Service User Representative; London Hearing Voices Project Manager
Peter Woodhams
Carer Representative
Norman Young
Nurse Consultant, Cardiff and Vale UHB & Cardiff University

TABLE OF CONTENTS
1 Preface .......................................................................................................................................... 8
1.1 National clinical guidelines ................................................................................................... 9
1.2 The national psychosis and schizophrenia guideline ........................................................... 11
2 Psychosis and schizophrenia in adults ................................................................................ 14

2.1 The disorder ......................................................................................................................... 14
2.2 Assessment, engagement, consent and the therapeutic alliance.......................................... 26
2.3 Language and stigma .......................................................................................................... 27
2.4 Issues for families, carers and friends .................................................................................. 28
2.5 Treatment and management of psychosis and schizophrenia in the NHS .......................... 29
2.6 Economic cost ...................................................................................................................... 40
3 Methods used to develop this guideline ............................................................................. 43

3.1 Overview ............................................................................................................................. 43
3.2 The scope.............................................................................................................................. 43
3.3 The Guideline Development Group..................................................................................... 44
3.4 Review questions ................................................................................................................. 45
3.5 Clinical review methods ...................................................................................................... 47
3.6 Health economics methods................................................................................................... 58
3.7 Linking evidence to recommendations................................................................................. 62
3.8 Stakeholder contributions .................................................................................................... 63
3.9 Validation of the guideline .................................................................................................. 64
4 Carers’ experience .................................................................................................................... 65

4.1 Introduction......................................................................................................................... 65
4.2 Carers’ experience (qualitative review) ............................................................................... 66
4.3 Interventions to improve carers’ experience ........................................................................ 82
4.4 Health economics evidence .................................................................................................. 96
4.5 Linking evidence to recommendations................................................................................. 97
4.6 Recommendations .............................................................................................................. 101
5 Preventing psychosis and schizophrenia: treatment of at risk mental states ............. 102
5.1 Introduction....................................................................................................................... 102
5.2 Clinical review protocol for at risk mental states for psychosis and schizophrenia........... 103
5.3 Pharmacological interventions .......................................................................................... 105
5.4 Dietary interventions ........................................................................................................ 116
5.5 Psychosocial interventions ................................................................................................ 119

5.6 Health economic evidence .................................................................................................. 129
5.7 Linking evidence to recommendations............................................................................... 133
6 Access and engagement......................................................................................................... 138

6.1 Introduction....................................................................................................................... 138
6.2 Access and engagement to service-level interventions ...................................................... 138
7 Interventions to promote physical health in adults......................................................... 157

7.1 Introduction....................................................................................................................... 157
7.2 Behavioural interventions to promote physical activity and healthy eating ..................... 157
7.3 Interventions for smoking cessation and reduction........................................................... 172
8 Peer-provided and self-management interventions ........................................................ 184

8.1 Introduction....................................................................................................................... 184
8.2 Peer-provided interventions .............................................................................................. 184
8.3 Self-management interventions......................................................................................... 195
9 Psychological therapy and psychosocial interventions .................................................. 208

9.1 Introduction....................................................................................................................... 208
9.2 Adherence therapy ............................................................................................................. 212
9.3 Arts therapies .................................................................................................................... 216
9.4 Cognitive behavioural therapy .......................................................................................... 221
9.5 Cognitive remediation ....................................................................................................... 242
9.6 Counselling and supportive therapy ................................................................................. 250
9.7 Family intervention ........................................................................................................... 256
9.8 Psychodynamic and psychoanalytical therapies ................................................................ 277
9.9 Psychoeducation ................................................................................................................ 281
9.10 Social skills training .......................................................................................................... 286
9.11 Psychological management of trauma in psychosis and schizophrenia............................. 293
9.12 **2009**Recommendations (across all treatments) ........................................................... 300
10 Pharmacological interventions in the treatment and management of schizophrenia.....

.................................................................................................................................................... 301
10.1 Introduction....................................................................................................................... 302
10.2 Initial treatment with antipsychotic medication ............................................................... 306
10.3 Oral antipsychotics in the treatment of the acute episode ................................................. 311
10.4 Promoting recovery in people with schizophrenia that are in remission – pharmacological
relapse prevention ........................................................................................................................... 320

10.5 Promoting recovery in people with schizophrenia whose illness has not responded
adequately to treatment .................................................................................................................. 327
10.6 Treatment with depot/ long-acting injectable antipsychotic medication .......................... 347
10.7 Side effects of antipsychotic medication............................................................................. 352
10.8 Effectiveness of antipsychotic medication.......................................................................... 357
10.9 Health economics ............................................................................................................... 359
11 Economic model - cost effectiveness of pharmacological interventions for people

with schizophrenia ......................................................................................................................... 386
11.1 Introduction....................................................................................................................... 386
11.2 Economic modelling methods ............................................................................................ 387
11.3 Results ............................................................................................................................... 440
11.4 Discussion of findings - limitations of the analysis .......................................................... 449
11.5 Conclusions ....................................................................................................................... 455
12 Teams and service-level interventions ............................................................................... 457

12.1 Introduction....................................................................................................................... 457
12.2 Interface between primary and secondary care.................................................................. 458
12.3 Non-acute Community mental healthcare ........................................................................ 468
12.4 Alternatives to acute admission ........................................................................................ 506
13 Vocational rehabilitation ...................................................................................................... 531

13.1 Introduction....................................................................................................................... 531
13.2 Clinical evidence review – vocational rehabilitation interventions................................... 532
13.3 Health economics evidence ................................................................................................ 561
14 Summary of recommendations ............................................................................................ 581

14.1 Care across all phases ........................................................................................................ 581
14.2 Preventing psychosis ......................................................................................................... 584
14.3 First episode psychosis....................................................................................................... 585
14.4 Subsequent acute episodes of psychosis or schizophrenia and referral in crisis ................ 591
14.5 Promoting recovery and possible future care .................................................................... 593
14.6 Research recommendations................................................................................................ 598
15 References ................................................................................................................................ 601

ACKNOWLEDGEMENTS
The Guideline Development Group (GDG) and the National Collaborating Centre
for Mental Health (NCCMH) review team would like to thank the following people:
Those who acted as advisors on specialist topics or have contributed to the process
by meeting the Guideline Development Group:
Victoria Bird, King’s College London
Brynmor Lloyd-Evans, Mental Health Sciences Unit, University College London
Alyssa Milton, Mental Health Sciences Unit, University College London
Daniel Tsoi, University of Sheffield
Sophia Winterbourne, London School of Economics
Those who conducted a review on behalf of the GDG:
Brynmor Lloyd-Evans, Mental Health Sciences Unit, University College London
Alyssa Milton, Mental Health Sciences Unit, University College London
Luke Sheridan Rains, Mental Health Sciences Unit, University College London
Research assistance
Saima Ali
Editorial assistance
Nuala Ernest

1 PREFACE
This guideline was first published in December 2002 (NCCMH, 2003; NICE, 2002b)
(referred to as the ‘2002 guideline’) and updated in 2009 (NCCMH, 2010 [full
guideline]; NICE, 2009d) (referred to as the ‘2009 guideline’). The 2009 guideline
updated most areas of the 2002 guideline, except for some service-level interventions
and the use of rapid tranquillisation. This second update (referred to as the ‘2014
guideline’) reviews the areas of service-level interventions that were not updated in
the 2009 guideline such as peer support and self-management interventions,
vocational rehabilitation and teams and service-level interventions that encompass
community-based interventions and alternatives to acute admission. In addition, the
2014 guideline provides a new review of carers’ experience and physical healthcare.
Given the change to the title (Psychosis and Schizophrenia rather than Schizophrenia),
the 2014 guideline also incorporates a review on at risk mental states, and in the
updated sections of the 2014 guideline, including the recommendations, the term
‘psychosis and schizophrenia’ is used rather than ‘schizophrenia’. The chapter on
experience of care in the 2009 guideline has been removed because it was
superseded by Service User Experience in Adult Mental Health (NICE clinical guidance
136 (2012 [full guideline])). For a full version of the 2009 guideline see Appendix 27.
See Appendix 1 for more details on the scope of the 2014 guideline. Sections of the
guideline where the evidence has not been updated since 2009 are marked by
asterisks and the date (**2009**_**2009**). Sections where the evidence has not been
updated since the 2002 are marked by asterisks and the date (**2002**-**2002**).
This guideline has been developed to advise on the treatment and management of
psychosis and schizophrenia in adults. The guideline recommendations have been
developed by a multidisciplinary team of healthcare professionals, people with
psychosis and schizophrenia, their carers and guideline methodologists after careful
consideration of the best available evidence. It is intended that the guideline will be
useful to clinicians and service commissioners in providing and planning high-
quality care for people with psychosis and schizophrenia while also emphasising the
importance of the experience of care for people with psychosis and schizophrenia
and their carers (see Appendix 1 for more details on the scope of the guideline).
Although the evidence base is rapidly expanding, there are a number of major gaps
and future revisions of this guideline will incorporate new scientific evidence as it
develops. The guideline makes a number of research recommendations specifically
to address gaps in the evidence base. In the meantime, it is hoped that the guideline
will assist clinicians, and people with psychosis and schizophrenia and their carers
by identifying the merits of particular treatment approaches where the evidence
from research and clinical experience exists.

1.1 NATIONAL CLINICAL GUIDELINES
1.1.1 What are clinical guidelines?
Clinical guidelines are ‘systematically developed statements that assist clinicians and
service users in making decisions about appropriate treatment for specific
conditions’ (Mann, 1996). They are derived from the best available research
evidence, using predetermined and systematic methods to identify and evaluate the
evidence relating to the specific condition in question. Where evidence is lacking, the
guidelines incorporate statements and recommendations based upon the consensus
statements developed by the Guideline Development Group (GDG).
Clinical guidelines are intended to improve the process and outcomes of healthcare
in a number of different ways. They can:
• provide up-to-date evidence-based recommendations for the management of

conditions and disorders by healthcare professionals
• be used as the basis to set standards to assess the practice of healthcare
professionals
• form the basis for education and training of healthcare professionals
• assist service users and their carers in making informed decisions about their
treatment and care
• improve communication between healthcare professionals, service users and
their carers
• help identify priority areas for further research.
1.1.2 Uses and limitation of clinical guidelines

Guidelines are not a substitute for professional knowledge and clinical judgement.
They can be limited in their usefulness and applicability by a number of different
factors: the availability of high-quality research evidence, the quality of the
methodology used in the development of the guideline, the generalisability of
research findings and the uniqueness of individuals.
Although the quality of research in this field is variable, the methodology used here
reflects current international understanding on the appropriate practice for guideline
development (Appraisal of Guidelines for Research and Evaluation Instrument
[AGREE]; www.agreetrust.org; AGREE Collaboration (2003)), ensuring the
collection and selection of the best research evidence available and the systematic
generation of treatment recommendations applicable to the majority of people with
psychosis and schizophrenia. However, there will always be some people for whom
and situations for which clinical guideline recommendations are not readily
applicable. This guideline does not, therefore, override the individual responsibility
of healthcare professionals to make appropriate decisions in the circumstances of the
individual, in consultation with the person with psychosis and schizophrenia or
their carer.

In addition to the clinical evidence, cost-effectiveness information, where available,
is taken into account in the generation of statements and recommendations of the
clinical guidelines. While national guidelines are concerned with clinical and cost
effectiveness, issues of affordability and implementation costs are to be determined
by the National Health Service (NHS).
In using guidelines, it is important to remember that the absence of empirical

evidence for the effectiveness of a particular intervention is not the same as evidence
for ineffectiveness. In addition, and of particular relevance in mental health,
evidence-based treatments are often delivered within the context of an overall
treatment programme including a range of activities, the purpose of which may be to
help engage the person and provide an appropriate context for the delivery of
specific interventions. It is important to maintain and enhance the service context in
which these interventions are delivered, otherwise the specific benefits of effective
interventions will be lost. Indeed, the importance of organising care in order to
support and encourage a good therapeutic relationship is at times as important as
the specific treatments offered.
1.1.3 Why develop national guidelines?

The National Institute for Health and Care Excellence (NICE) was established as a
Special Health Authority for England and Wales in 1999, with a remit to provide a
single source of authoritative and reliable guidance for service users, professionals
and the public. NICE guidance aims to improve standards of care, diminish
unacceptable variations in the provision and quality of care across the NHS, and
ensure that the health service is person-centred. All guidance is developed in a
transparent and collaborative manner, using the best available evidence and
involving all relevant stakeholders.
NICE generates guidance in a number of different ways, three of which are relevant
here. First, national guidance is produced by the Technology Appraisal Committee
to give robust advice about a particular treatment, intervention, procedure or other
health technology. Second, NICE commissions public health intervention guidance
focused on types of activity (interventions) that help to reduce people’s risk of
developing a disease or condition, or help to promote or maintain a healthy lifestyle.
Third, NICE commissions the production of national clinical guidelines focused
upon the overall treatment and management of a specific condition. To enable this
latter development, NICE has established four National Collaborating Centres in
conjunction with a range of professional organisations involved in healthcare.
1.1.4 From national clinical guidelines to local protocols

Once a national guideline has been published and disseminated, local healthcare
groups will be expected to produce a plan and identify resources for
implementation, along with appropriate timetables. Subsequently, a
multidisciplinary group involving commissioners of healthcare, primary care and
specialist mental health professionals, service users and carers should undertake the
translation of the implementation plan into local protocols, taking into account both

the recommendations set out in this guideline and the priorities set in the National
Service Framework for Mental Health (Department of Health, 1999) and related
documentation. The nature and pace of the local plan will reflect local healthcare
needs and the nature of existing services; full implementation may take a
considerable time, especially where substantial training needs are identified.
1.1.5 Auditing the implementation of clinical guidelines

This guideline identifies key areas of clinical practice and service delivery for local
and national audit. Although the generation of audit standards is an important and
necessary step in the implementation of this guidance, a more broadly-based
implementation strategy will be developed. Nevertheless, it should be noted that the
Care Quality Commission will monitor the extent to which commissioners and
providers of health and social care have implemented these guidelines.
1.2 THE NATIONAL PSYCHOSIS AND SCHIZOPHRENIA

GUIDELINE
1.2.1 Who has developed this guideline?
This guideline has been commissioned by NICE and developed within the National
Collaborating Centre for Mental Health (NCCMH). The NCCMH is a collaboration
of the professional organisations involved in the field of mental health, national
service user and carer organisations, a number of academic institutions and NICE.
The NCCMH is funded by NICE and is led by a partnership between the Royal
College of Psychiatrists and the British Psychological Society’s Centre for Outcomes
Research and Effectiveness, based at University College London.
The GDG was convened by the NCCMH and supported by funding from NICE. The
GDG included people with psychosis and schizophrenia and carers, and
professionals from psychosis and schizophrenia psychiatry, clinical psychology,
general practice, occupational therapy, nursing, psychiatric pharmacy, and the
private and voluntary sectors.
Staff from the NCCMH provided leadership and support throughout the process of
guideline development, undertaking systematic searches, information retrieval,
appraisal and systematic review of the evidence. Members of the GDG received
training in the process of guideline development from NCCMH staff, and the service
users and carers received training and support from the NICE Patient and Public
Involvement Programme. The NICE Guidelines Technical Adviser provided advice
and assistance regarding aspects of the guideline development process.
All GDG members made formal declarations of interest at the outset, which were
updated at every GDG meeting. The GDG met a total of eleven times throughout the
process of guideline development. The GDG was supported by the NCCMH
technical team, with additional expert advice from special advisers where needed.
The group oversaw the production and synthesis of research evidence before

presentation. All statements and recommendations in this guideline have been
generated and agreed by the whole GDG.
1.2.2 For whom is this guideline intended?

This guideline will be relevant for adults with psychosis and schizophrenia and
covers the care provided by primary, community, secondary, tertiary and other
healthcare professionals who have direct contact with, and make decisions
concerning the care of, adults with psychosis and schizophrenia.
The guideline will also be relevant to the work, but will not cover the practice, of
those in:
• occupational health services
• social services
• the independent sector
• Other professional bodies/ group who have direct contact with people
with psychosis or schizophrenia.
1.2.3 Specific aims of this guideline

The guideline makes recommendations for the treatment and management of
psychosis and schizophrenia. It aims to:
• improve access and engagement with treatment and services for people with
psychosis and schizophrenia
• evaluate the role of specific psychological, psychosocial and pharmacological
interventions in the treatment of psychosis and schizophrenia
• evaluate the role of psychological and psychosocial interventions in
combination with pharmacological interventions in the treatment of
• evaluate the role of specific service-level interventions for people with
• integrate the above to provide best-practice advice on the care of individuals
throughout the course of their psychosis and schizophrenia
• promote the implementation of best clinical practice through the development
of recommendations tailored to the requirements of the NHS in England and
Wales.
1.2.4 The structure of this guideline

The guideline is divided into chapters, each covering a set of related topics. The first
three chapters provide a summary of the clinical practice and research
recommendations, and a general introduction to guidelines and to the methods used
to develop them. For the methods used in 2009 relating to chapters 6, 9, 10 and 11 see
Appendix 11. Chapter 4 to Chapter 13 provide the evidence that underpins the
recommendations about the treatment and management of psychosis and
schizophrenia.

Each evidence chapter begins with a statement about whether the chapter has been
updated and a general introduction to the topic that sets the recommendations in
context. Depending on the nature of the evidence, narrative reviews or meta-
analyses were conducted, and the structure of the chapters varies accordingly.
Where appropriate, details about current practice, the evidence base and any
research limitations are provided. Where meta-analyses were conducted,
information is given about both the interventions included and the studies
considered for review. Clinical summaries are then used to summarise the evidence
presented. Finally, recommendations related to each topic are presented at the end of
each evidence review or at the end of the chapter, as appropriate. In the separate
appendix files, full details about the included and excluded studies for the 2014
guideline can be found in Appendix 15 (for evidence reviewed in 2009 see Appendix
22). Where meta-analyses were conducted, the data for the 2014 guideline are
presented using forest plots in Appendix 16 (for evidence reviewed in 2009 see
Appendix 23) (see Text Box 1 for details).
Text Box 1: Appendices in a separate file
2014 Search strategies for the identification of clinical studies Appendix 13

2014 Search strategies for the identification of health economics evidence Appendix 14
2014 Study characteristics for quantitative studies Appendix 15a
2014 Study characteristics for qualitative studies Appendix 15b
2014 Clinical evidence forest plots Appendix 16
2014 GRADE evidence profiles (clinical and health economic) Appendix 17
2014 Health economic evidence- completed methodology checklists Appendix 18
2014 Health economic evidence- evidence tables of published studies Appendix 19
2009 Search strategies for clinical evidence Appendix 20
2009 Clinical review and clinical questions Appendix 21
2009 Study characteristics for clinical evidence Appendix 22
2009 Clinical evidence forest plots and/ or data tables Appendix 23
2009 Search strategies for the identification of health economics evidence Appendix 24
2009 Search strategies for the identification for economic studies Appendix 25
2009 Winbugs codes used for mixed treatment comparisons in the Appendix 26
economic model of pharmacological treatments for relapse prevention
2009 The full Schizophrenia in adults guideline Appendix 27
2009 Health economics checklist Appendix 28
In the event that amendments or minor updates need to be made to the guideline,
please check the NCCMH website (nccmh.org.uk), where these will be listed and a
corrected PDF file available to download.

2 PSYCHOSIS AND SCHIZOPHRENIA
IN ADULTS
This guideline is concerned with the treatment and management of the non-specific
diagnosis of psychosis and with the more specific diagnosis of schizophrenia in
adults, as defined in the International Classification of Diseases, 10th Revision (ICD-10)
(World Health Organization, 1992), in the community, in hospital and in prison. The
term ‘psychosis’ covers a set of related conditions, of which the commonest is
schizophrenia, and includes schizoaffective disorder, schizophreniform disorder,
delusional disorder and the so-called non-affective psychoses. This guideline does
not address the treatment and management of other psychotic disorders, such as
bipolar disorder and unipolar psychotic depression, or psychosis and schizophrenia
in children and young people, because they are covered by other NICE guidelines.
2.1 THE DISORDER

2.1.1 Symptoms and presentation
Psychosis and the specific diagnosis of schizophrenia represent a major psychiatric
disorder (or cluster of disorders) in which a person’s perceptions, thoughts, mood
and behaviour are significantly altered. Individuals who develop psychosis or
schizophrenia will each have their own unique combination of symptoms and
experiences, which will vary depending on their particular circumstances.
In the decade since the first NICE guideline on schizophrenia (2002b), there has been
a considerable shift in understanding the complexity of psychosis and
schizophrenia, with a greater appreciation of the role of affect in non-affective
psychoses, and in the continua of processes that underlie the disorders. Current
understanding is ‘still limited by the substantial clinical, pathological and etiological
heterogeneity of schizophrenia and its blurred boundaries with several other
psychiatric disorders, leading to a ‘fuzzy cluster’ or overlapping syndromes, thereby
reducing the content, discriminant and predictive validity of a unitary construct’
(Keshavan et al., 2011) .
Typically, there will be a ‘prodromal’ period often characterised by some

deterioration in personal functioning. Difficulties may include memory and attention
problems, social withdrawal, unusual and uncharacteristic behaviour, disturbed
communication and affect, unusual perceptual experiences, which are accompanied
by bizarre ideas, poor personal hygiene, and reduced interest in day-to-day
activities. During this prodromal period, people with psychosis often feel that their
world has changed, but their interpretation of this change may not be shared by
others. Relatives and friends usually notice this as changes ‘in themselves’. The
changes may affect the person’s ability to study, to hold down employment, or
maintain relationships; they may become increasingly isolated.

This prodromal period is typically followed by an acute phase marked by positive
symptoms, such as hallucinations (hearing, seeing or feeling things that others do
not), delusions (markedly unusual or bizarre ideas), behavioural disturbances such
as agitation and distress, and disorders of thinking so that speech becomes muddled
and hard to understand. If these acute problems resolve, usually after some
treatment, the positive symptoms may disappear or reduce, but it is common for
negative symptoms such as poor motivation, poor self-care and poor memory and
attention to remain problematic. This may interfere with the person’s ability to
return to study, to work and to manage their day to day activities.
Affective dysfunction and comorbidities are now recognised to be highly prevalent

in people with psychosis and schizophrenia; indeed those studies that have analysed
the symptom structure of psychotic experience, all include a dimension of
depression and related symptoms, even in 'non-affective' diagnoses (Russo et al.,
2013). Over 90% of individuals with first episode psychosis report depression in the
prodrome, during the acute episode, or in the year following recovery of positive
symptoms (Upthegrove et al., 2010). Social anxiety disorder that is not attributable to
paranoia is present in up to a third of individuals with psychosis and schizophrenia,
with similar figures for post-traumatic stress disorder (PTSD). While figures for
social anxiety disorder and PTSD remain constant across phases, depression tends to
peak during the prodrome and in acute psychosis but declines to about one-third
following recovery. It has been shown that there are several pathways to emotional
dysfunction in psychosis, including the common background of social risk factors for
both psychosis and depression and as a psychological reaction to the diagnosis itself
(Birchwood, 2003).
People vary considerably in their pattern of symptoms and problems and in the
resulting course of any remaining difficulties. While most people will recover from
the initial acute phase, only 14 to 20% will recover fully. Others will improve but
have recurrent episodes or relapses, the timing of which are related to stress,
adversity, social isolation and poor take-up of treatments. Thus some people have
disturbing experiences only briefly, whereas others will live with them for months or
years. In the longer term (up to 15 years) over half of those diagnosed will have
episodic rather than continuous difficulties. As Harrow and colleagues (2005) have
observed, ‘some of these intervals of recovery will appear spontaneously and may be
tied to individual factors, such as resilience.’
2.1.2 At risk mental states

In recent years there has been a growing emphasis on early detection and
intervention in order to delay or possibly prevent the onset of psychosis and
schizophrenia. This focus on very early intervention and prevention has stimulated
an interest in identifying, and potentially intervening in, the so-called ‘at risk mental
states’ (or prodrome) which may precede the onset of the disorder.

At risk or ‘ultra-high risk’ mental states, are characterised by help-seeking behaviour
and the presence of attenuated (subclinical) positive psychotic symptoms, brief
limited intermittent psychotic symptoms or a combination of genetic risk indicators,
such as the presence of schizotypal disorder, with recent functional deterioration.
Although the risk for schizophrenia emerging over a 12-month period appears to be
increased (between one in five to one in ten may be expected to develop a
schizophrenic disorder (Ruhrmann et al., 2010)), it remains the case that prediction
of schizophrenia based on at risk or ultra-high risk mental states is modest given that
the majority of those identified do not become psychotic. Furthermore, most people
identified with at risk mental states have a mixture of other mental health problems
(for example, depression, anxiety, substance-use disorders or emerging personality
disorder) requiring a range of targeted interventions. In addition, the potential use of
a clinical label that conveys a future risk of psychosis or schizophrenia raises ethical
issues and may itself be perceived as stigmatising. It may be that at risk or ultra-high
risk mental states are best viewed as a dimension rather than a diagnostic category,
including at one extreme people with non-specific symptoms and at the other those
on the cusp of psychosis. Finally, given the low rate of transition to psychosis, any
interventions used must benefit (and not harm) the majority of people (false
positives) who do not develop psychosis.
2.1.3 Impairment and disability

Although the problems and experiences associated with psychosis and
schizophrenia are often distressing, the effects of the disorder can be pervasive. A
significant number of people continue to experience long-term impairments, and as
a result psychosis and schizophrenia can have a considerable effect on people’s
personal, social and occupational lives. A European study of six countries found that
over 80% of adults with this diagnosis had some persistent problems with social
functioning, though not all of them were severe. The best predictor of poorer
functioning in the long term was poor functioning in the first 3 years post-diagnosis
(Wiersma et al., 2000), particularly for unemployment, which was linked to duration
of untreated psychosis and increased negative symptoms (Turner et al., 2009).
Current estimates of employment for people with schizophrenia are 5 to 15% with an
average of 8% (Schizophrenia Commission, 2012), which is significantly less than the
general population (of which 71 % are currently employed).
The disabilities experienced by people with psychosis and schizophrenia are not
solely the result of recurrent episodes or continuing symptoms. Unpleasant side
effects of treatment, social adversity and isolation, poverty and homelessness also
play a part. These difficulties are not made any easier by the continuing prejudice,
stigma and social exclusion associated with the diagnosis (Sartorius, 2002;
Thornicroft, 2006).
Worldwide, it has been estimated that schizophrenia falls into the top fifteen medical
disorders causing disability (Murray et al., 2013). Mortality among people with
schizophrenia is approximately 50% above that of the general population. This is
partly as a result of an increased incidence of suicide (an approximate lifetime risk of

5% (Hor & Taylor, 2010)) and violent death, and partly because of an increased risk
of a wide range of physical health problems.
Cardiovascular events have been found to be the largest single contributor, with
illnesses associated with obesity, metabolic aberrations, smoking, alcohol, lack of
exercise, poor diet and diabetes, making significant contributions (von Hausswolff-
Juhlin et al., 2009). The precise extent to which high mortality and disability rates
are, at least in part, a result of some of the medications prescribed for schizophrenia
is still not clear (Weinmann et al., 2009). Difficulties experienced by people with
mental health problems in accessing general medical services in both primary and
secondary care continue to contribute to reduced life expectancy (Lawrence &
Kisely, 2010). Recent work indicates that young Caribbean and African men, and
middle-aged women from diverse ethnic or cultural backgrounds, are at higher risk
of suicide, and that this may be because of differences in symptom presentation and
conventional risk-factor profiles across ethnic groups (Bhui & McKenzie, 2008).
2.1.4 Prognosis, course and recovery

Historically, many psychiatrists and other healthcare professionals have taken a
pessimistic view of the prognosis for schizophrenia, regarding it as a severe,
intractable and often deteriorating lifelong illness. This negative view has failed to
find confirmation from long-term follow-up studies, which have demonstrated
considerable variations in long-term outcome. While it is estimated that around
three-quarters of people with schizophrenia will experience recurrent relapse and
some continued disability (Brown et al., 2010), the findings of follow-up studies over
periods of 20 to 40 years suggest that there is a moderately good long-term global
outcome in over half of people with schizophrenia, with a smaller proportion having
extended periods of remission of symptoms without further relapses (Banham &
Gilbody, 2010; Harrison et al., 2001; Jobe & Harrow, 2005). It should also be noted
that some people who never experience complete recovery from their experiences
nonetheless manage to sustain an acceptable quality of life if given adequate support
and help.
The early stages of psychosis and schizophrenia are often characterised by repeated
exacerbation of symptoms such as hallucinations and delusions and disturbed
behaviour. While a high proportion respond to initial treatment with antipsychotic
medication, around 80% will relapse within 5 years of a treated first episode, which
is partly explained by discontinuation of medication (Brown et al., 2010).
Research has suggested that delayed access to mental health services and treatment
in early psychosis and schizophrenia – often referred to as the duration of untreated
psychosis – is associated with slower or less complete recovery, and increased risk of
relapse and poorer outcome in subsequent years (Bottlender et al., 2003; Harrigan et
al., 2003; Robinson et al., 1999).
In the UK and other countries early intervention in psychosis teams have been
introduced with an aim of reducing delay to treatment in order to try to improve

outcomes. In the longer term, the factors that influence the differential recovery from
psychosis and schizophrenia are not well known. But recovery may happen at any
time, even after many years (Harrison et al., 2001).
A number of social and economic factors appear to affect the course of psychosis and
schizophrenia. For example, in developed countries it is well established that
psychosis and schizophrenia is more common in lower socioeconomic groups.
However, this appears to be partly reversed in some developing countries (Jablensky
et al., 1992), suggesting that the relationship between incidence, recovery rates, and
cultural and economic factors is more complex than a simple correspondence with
socioeconomic deprivation (Warner, 1994). There is some evidence that clinical
outcomes are worse in Europe than in East Asia, Latin America, North Africa and
the Middle East (Haro et al., 2011).
The risk factors for developing psychosis and schizophrenia and the acceptability of
interventions and the uptake of treatments have been shown to vary across ethnic
groups. Although the focus in the UK has been on African and Caribbean
populations, some evidence suggests other ethnic groups and migrants in general
may be at risk; social risk factors may be expressed through an ethnic group, rather
than being an intrinsic risk for that ethnic groups per se. However, the different
pattern of service use, access to services and perceived benefits across ethnic groups
is a cause of concern among service users.
The effects of psychosis and schizophrenia on a person’s life experience and

opportunities are considerable; service users and carers need help and support to
deal with their future and to cope with any changes that may happen.
2.1.5 Diagnosis
Although a full discussion of the diagnoses of psychosis and schizophrenia is
outside the scope of this guideline, some specific issues are discussed here to provide
context.
ICD-10 (World Health Organisation, 1992) describes symptom clusters necessary for
the diagnosis of different subtypes of schizophrenia. For some subtypes, ICD-10
requires that clear psychotic symptoms be present for only 1 month, with any period
of non-specific impairment or attenuated (prodromal) symptoms that may precede
an acute episode not counted. In ICD-10, evidence of deteriorating and impaired
functioning in addition to persistent psychotic symptoms is essential for a diagnosis.
Isolated psychotic symptoms (typically auditory hallucinations) without functional
impairment are surprisingly common in both the general population (van Os et al.,
2009) and people with emotional disorders such as anxiety and depression
(Varghese et al., 2011); such experiences should not be confused with a diagnosis of a
psychotic disorder or schizophrenia.
The experience of a psychotic disorder challenges an individual’s fundamental

assumption that they can rely upon the reality of their thoughts and perceptions.

This is often both frightening and emotionally painful for both the service user and
for those close to them. For this experience then to be classified as a disorder and to
acquire a diagnostic label may either be helpful in facilitating understanding or may
be experienced as yet a further assault upon one’s identity and integrity.
Professionals need to be aware of both the positive and negative impacts of
discussing a diagnosis (Pitt et al., 2009); positive aspects can include naming the
problem and providing a means of access to appropriate help and support; negative
aspects can include ‘labelling’ the person, stigma and discrimination and
disempowerment. The toxicity of the label of ‘schizophrenia’ has led to calls to
abandon the concept altogether (Bentall et al., 1988) or to rename the condition
(Kingdon et al., 2007). This has led to some professionals and user/carer groups
questioning the usefulness of diagnosis and instead preferring to emphasise a
narrative or psychological formulation of an individual’s experiences. There is some
evidence that psychosocial explanations of psychosis are less associated with stigma,
desire for social distance and perceptions of dangerousness and uncontrollability
than biomedical explanations (such as a diagnosis of an illness) in the general public
(Read et al., 2006), healthcare professionals (Lincoln et al., 2008) and service users
(Wardle et al., In press).
The majority of people for whom a diagnosis of psychosis or schizophrenia is being

considered will be in their first episode of illness, although the literature on duration
of untreated psychosis would suggest some of these may have had psychotic
experiences for many years (Marshall et al., 2005). The future course and diagnostic
stability of an initial psychotic episode shows much variation, with a sizable
proportion (approximately 20%) only having one episode (Rosen & Garety, 2005). In
addition to a lack of predictive validity regarding course and outcome, there are also
significant problems with the reliability of the diagnosis (Bentall, 1993). It is
recognised that accurate diagnosis is particularly challenging in the early phases of
psychosis, which has led early intervention for psychosis services to ‘embrace
diagnostic uncertainty’ (Singh & Fisher, 2005).
For all of the above reasons, the less specific umbrella term ‘psychosis’ has, therefore,
found increasing favour in some professionals and some user/carer groups.
2.1.6 Physical health

The association between psychosis/schizophrenia and poor physical health is well
established (Marder et al., 2003). Males with schizophrenia die 20 years earlier and
females 15 years earlier than the general population (Wahlbeck et al., 2011). About a
fifth of premature deaths arise from suicide and accidents but most are accounted for
by physical disorders (Brown et al., 2010; Lawrence et al., 2013; Saha et al., 2007),
which include cardiovascular disorders (for example, coronary heart disease,
peripheral vascular disease and stroke), metabolic disorders such as diabetes
mellitus, chronic obstructive pulmonary disease, certain cancers and infectious
disorders such as HIV, hepatitis C and tuberculosis (Leucht et al., 2007). And
although not life-threatening, difficulties such as sexual dysfunction, dental caries

(Friedlander & Marder, 2002), constipation and nocturnal enuresis (Barnes et al.,
2012) can be distressing and socially isolating.
While much of the increased burden of poor physical health can be explained by the
nature of psychosis and schizophrenia and side effects of treatment, this
‘undoubtedly also results from the unsatisfactory organization of health services,
from the attitudes of medical doctors, and the social stigma ascribed to the
schizophrenic patients’(Leucht et al., 2007). Despite having two to three times the
likelihood of developing diabetes mellitus compared with the general population,
this condition often goes unrecognised in people with schizophrenia. In a study from
the Maudsley hospital in London, a chart review indicated that 39 (6.1%) of 606
inpatients had diabetes or impaired glucose tolerance; when undiagnosed
individuals were formally tested for diabetes by a fasting blood glucose
measurement, a further 16% were discovered to have either diabetes or impaired
fasting glucose (Taylor et al., 2005). A European study screening people with
schizophrenia who were not known to have diabetes, discovered 10% had type 2
diabetes and 38% were at high risk of type 2 diabetes; this population’s average age
was only 38 years (Manu et al., 2012).
A recent Scottish study of 314 general practices compared the nature and extent of
physical health comorbidities between 9,677 people with psychosis and
schizophrenia and 1,414,701 controls (Smith et al., 2013). Based on the presence of a
possible recorded diagnosis for 32 index physical conditions, the study found that
people with schizophrenia were more likely to experience multiple physical
comorbidities: higher rates of viral hepatitis, constipation and Parkinson’s disorder
but lower than expected rates of CVD. The authors concluded there was a systematic
under-recognition and under treatment of CVD in people with schizophrenia in
primary care, which might contribute to the substantial cardiovascular-related
morbidity and premature mortality observed in this group.
A similar picture of late recognition and under-treatment is apparent for cancer,

although intriguingly a recent study from Sweden revealed decreased incidences of
certain cancers in people with schizophrenia and their unaffected relatives (Ji et al.,
2013). The authors suggested that familiar/genetic factors contributing to
schizophrenia may protect against the development of cancer; this protective effect
did not hold for breast, cervical and endometrial cancers, where rates were higher in
women with schizophrenia. Nevertheless, even with these protective factors towards
certain cancers, people with schizophrenia are more likely to have metastases at
diagnosis and less likely to receive specialised interventions (Kisely et al., 2013),
which explains why they are still more likely to die prematurely from cancer than
the general population (Bushe et al., 2010).
The impact of cardiovascular diseases

The reduction in cardiovascular morbidity and mortality seen in the general
population over the last 2 decades has not been seen in people with severe mental
illness in whom CVD remains the single biggest contributor to premature death

(Saha et al., 2007). Moreover, there is a widening mortality gap for people with
schizophrenia mainly as a result of higher relative rates of CVD compared with the
general population (Brown et al., 2010; Hennekens et al., 2005; Lawrence et al., 2003;
Osborn et al., 2007a).
CVD may result from the body’s response to persisting stress/distress, potential
genetic vulnerabilities, lifestyle issues (for example, tobacco use, diet, sedentariness,
poverty and exclusion) and psychiatric medication (De Hert et al., 2009b). The
tendency for metabolic risks to cluster together is conceptualised within the
metabolic syndrome, reliably predicting future CVD, diabetes and premature death;
the presence of central obesity is a core factor, usually combined with evidence of
impaired glucose handling, lipid abnormalities and hypertension (Alberti et al.,
2005). This is a significant problem for those with established schizophrenia (De Hert
et al., 2009b); for example, a Finnish cohort study revealed that by the age of 40
metabolic syndrome was four times more likely than in non-psychiatric populations
(Saari et al., 2005).
Antipsychotic medication
Antipsychotic medication may cause metabolic/endocrine abnormalities (for
example, weight gain, diabetes, lipid abnormalities and galactorrhoea), neurological
disorders (for example, tardive dyskinesia) and cardiac abnormalities (for example,
lengthened QT interval on electrocardiography) (American Diabetes Association et
al., 2004; Expert Group, 2004; Holt et al., 2005; Koro et al., 2002; Lieberman et al.,
2005; Lindenmayer et al., 2003; Nasrallah, 2003; Nasrallah, 2008; Saari et al., 2004;
Thakore, 2005). The effects of antipsychotics on CVD risk factors such as weight gain
and diabetes are examined in the sections below.
Weight gain, metabolic disturbance and antipsychotic medicines

The prevalence of obesity has increased dramatically in the general population over
the last 30 years, and has escalated even more rapidly in people with schizophrenia
(Homel et al., 2002). It seems likely that environmental changes have provoked these
increases in both populations but schizophrenia may also have disease-specific
effects, such as genetic susceptibility, that have additive or synergistic actions to
increase weight further. However the most important factor related to weight gain in
people with schizophrenia is the use of antipsychotics, which are among the most
obesogenic drugs. Moreover a causal link between antipsychotics and weight gain
appears certain (Foley & Morley, 2011; Kahn et al., 2008; Tarricone et al., 2010). This
is important because weight gain may lead to insulin resistance and other adverse
impacts such as dyslipidaemia, diabetes and hypertension. The true impact may
have been obscured by a lack of critical evaluation of weight gain specifically in
people never previously exposed to antipsychotics. Many of the antipsychotic trials
used short follow-up times observing older people with established illness, many of
whom may already have gained weight from previous antipsychotic exposure. In
contrast the European First Episode Schizophrenia Trial (EUFEST) (Kahn et al.,
2008), examining weight gain in a treatment-naïve group of people with a first
episode, found that the percentage of those gaining more than 7% of body weight

during the first year of treatment was 86% for olanzapine, 65% for quetiapine, 53%
for haloperidol and 37% for ziprasidone. Citing the findings of this study, Nasrallah
concluded that neither ‘first-generation’ antipsychotics, such as haloperidol, nor
drugs promoted as being metabolically benign ‘second-generation’ antipsychotics,
such as ziprasidone, could be regarded as exceptions to the generalisation that any
antipsychotic was capable of causing significant weight gain (Nasrallah, 2011). A
more recent EUFEST study also revealed that pre-treatment rates of metabolic
syndrome were no different from prevalence rates estimated in a general population
of similar age (Fleischhacker et al., 2012).
Underlining the differential impact of antipsychotics on a treatment-naïve

population, a recent systematic review concluded that antipsychotic-induced weight
gain had been underestimated three- to four-fold in those with first episode
psychosis (Alvarez-Jimenez et al., 2008). Indeed the majority of the weight gained
will have done so within the first 3 years of treatment (Addington et al., 2006).
Because first episode psychosis often commences when a person is in their late teens
and 20s (Kirkbride et al., 2006) the impact of antipsychotics may coincide with a
critical development phase. Although limited comparative data hampers
conclusions, younger people appear more vulnerable to side effects than older
people (weight gain, extrapyramidal symptoms, metabolic problems, prolactin
elevation and sedation (Kumra et al., 2008)). Risk of weight gain may also be more
likely in those with a low baseline weight (De Hert et al., 2009a). Not only can early
weight gain eventually lead to obesity-related metabolic and cardiac disorders, but it
may also restrict healthy physical activities as basic as walking, and lead to a lack of
self-worth and confidence to participate (Vancampfort et al., 2011). In addition, other
adverse effects such as hyperprolactinaemia (causing menstrual disturbances, sexual
dysfunction and galactorrhoea) (Fedorowicz & Fombonne, 2005) and movement
disorders can result in poor medicine concordance, which in turn may lead to this
vulnerable group of young people experiencing a cycle of relapse and disillusion
with services (Hack & Chow, 2001).
Lifestyle factors
Tobacco use
Smoking tobacco is more common in people with psychosis and schizophrenia than
the general population, even when variation in socioeconomic status is allowed for
(Brown et al., 1999; Osborn et al., 2006), with 59% already smoking at the onset of
psychosis (six times more frequently than age-matched peers without psychosis
(Myles et al., 2012)). Whereas average smoking rates in the UK have fallen in the
general population from around 40% in 1980 to 20% currently (Fidler et al., 2011),
rates for people with established schizophrenia remain around 70% (Brown et al.,
2010), and this group may also be less likely to receive smoking cessation advice
thereby missing out on effective prevention of a potent cause of premature death
(Duffy et al., 2012; Himelhoch & Daumit, 2003). Paradoxically rates of lung cancer

appear uninfluenced (Gulbinat et al., 1992; Harris & Barraclough, 1998; Jeste et al.,
1996; Osborn et al., 2007a).
Diet, nutrition and physical activity

Weight can increase rapidly in the early treatment phase not only because of the use
of antipsychotic medication, but also as a result of a diet that is frequently low in
fruit and vegetables and high in fat and sugar, lack of physical activity and impaired
motivation to change health behaviours.
Fewer than 30% of people with schizophrenia are regularly active compared with
62% of people without a serious mental illness (Lindamer et al., 2008), and fewer
than 25% undergo the recommended 150 minutes per week of at least moderate-
intensity aerobic activity (Faulkner et al., 2006). It may also be important to
acknowledge the risks of sedentariness on cardiovascular risk; a recent study of
healthy volunteers showed that minimal-intensity physical activity (standing and
walking) of longer duration improves insulin action and plasma lipids more than
shorter periods of moderate to vigorous exercise (cycling) in sedentary subjects
when energy expenditure is comparable (Duvivier et al., 2013).
2.1.7 Incidence and prevalence

Psychosis is relatively common mental illness, with schizophrenia being the most
common form of psychotic disorder. A review of the incidence of psychosis and
schizophrenia in England between 1950 and 2009 (Kirkbride et al., 2012) found a
pooled incidence of 31.7 per 100,000 for psychosis and of 15 per 100,000 for
schizophrenia. Rates varied according to gender and age group, with rates generally
reducing with age (although with a second peak in women starting in the mid to late
40s). Men under the age of 45 were found to have twice the rate of schizophrenia
than women, but there was no difference in its incidence after this age. The rate of
schizophrenia was found to be significantly higher in black Caribbean (RR: 5.6;
95%CI: 3.4, 9.2; I2=0.77) and black African (RR: 4.7; 95% CI: 3.3, 6.8; I2=0.47) migrants
and their descendants, compared with the baseline population. The incidence of
psychosis has been reported to vary from place to place with rates in south-east
London (55 per 100,000 person years) being more than twice those in both
Nottingham and Bristol (25 per 100,000 person years and 22 per 100,000 person
years, respectively) (Morgan et al., 2006).
The National Survey of Psychiatric Morbidity in the UK found a population

prevalence of probable psychotic disorder of 5 per 1000 in the age group 16 to 74
years (Singleton et al., 2003). Schizophrenia has a point prevalence averaging around
0.45% and a lifetime expectancy of 0.7%, although there is considerable variation in
different areas and a higher risk in urban environments (van Os et al., 2010).

2.1.8 Possible causes
It is known that there are a number of genetic and environmental risk factors for
developing psychosis and schizophrenia, but there remains uncertainty about how
these factors fit together to cause the disorder (Tandon et al., 2008).
Concerning genetic risks, having a close relative with psychosis or schizophrenia is

the biggest risk factor for developing a psychotic disorder (Gilmore, 2010). However,
while genetic risk is substantial, it is not due to a single ‘schizophrenia’ gene, but to
many genes, each of which makes a small contribution (Sullivan et al., 2003). Genetic
risk may also involve rare but important events such as deletions or duplications of
genes (The International Schizophrenia Consortium, 2008).
Genetic risks are not sufficient to explain why some people develop psychosis and
schizophrenia while others do not – for example, most people with psychosis and
schizophrenia do not have an affected relative. Therefore, there must also be
environmental risks, both biological and psychosocial. Potential biological risks
include: complications before or during birth (such as infections, poor nutrition
while in the womb, maternal stress or birth trauma) (Meli et al., 2012); cannabis use,
especially in adolescence (Arseneault et al., 2004; Moore et al., 2007); older paternal
age at birth (Miller et al., 2011) and seasonality of birth (Davies et al., 2003); and
exposure to the protozoan parasite toxoplasma gondii (Torrey et al., 2012). Potential
psychosocial risks include: urban birth and exposure to living in cities (Vassos et al.,
2012); childhood and adult adversity, including poor rearing environments, sexual,
physical and emotional abuse, neglect and bullying (Bebbington et al., 2004; van
Dam et al., 2012; Varese et al., 2012; Wahlberg et al., 1997); and migration, especially
when the migrants are from a developing country or a country where the majority of
the population is black (Cantor-Graae & Selten, 2005).
Several theories attempt to explain how genetic risks might fit together with
biological and psychosocial risks to cause psychotic disorders. None of these theories
are proven. One well established theory is the neurodevelopmental hypothesis
(Fatemi & Folsom, 2009), which proposes that some people have a vulnerability to
developing psychosis and schizophrenia that arises due to the interaction of genetic
and environmental risks around the time of birth. For example, some people might
have genes that increase the chances of complications before or during birth and/or
have other genes that make it difficult to replace or repair damaged nerve cells when
a complication occurs. The theory proposes that such people will sometimes acquire
subtle neurological injuries that are not immediately obvious during childhood.
However, as the child enters adolescence, these subtle injuries somehow disrupt the
normal changes in brain connectivity that occur in all teenagers. The end result is
that the affected person becomes particularly sensitive to developing psychosis in
the presence of some of the environmental risks (for example, cannabis use)
described above. There is evidence to support the neurodevelopmental hypothesis,
for example, some people who develop schizophrenia have unusual personality
traits (schizotypy) (Nelson et al., 2013), minor developmental delays (Jaaskelainen et
al., 2008; Welham et al., 2009) and subtle neurological signs (Neelam et al., 2011). On

the other hand, the theory is too broad to be easily proven; no specific neurological
injury has been pinpointed (although brain scans of some people who develop
schizophrenia show a range of abnormalities); and not all people who develop
schizophrenia have the signs described above. Moreover the theory does not readily
explain the contribution of several known psychosocial risks, such as urbanicity or
migration.
An alternative theory is that everyone carries some degree of vulnerability to

developing psychosis and schizophrenia and that the critical factor in many people
is not genes or subtle neurological injuries, but the timing, nature and degree of
exposure to environmental risks (van Os et al., 2009). Proponents of this theory point
to numerous studies illustrating that risks like urban living, poverty and child abuse
are highly predictive of later psychotic symptoms with or without a genetic risk
being present (Read et al., 2005). Perhaps psychological trauma in the early stages of
development can set up psychological vulnerabilities that can lead to psychosis in
later life in the face other environmental risks (van Os et al., 2010). In favour of this
theory is the discovery that isolated psychotic symptoms are common in the general
population, and that psychotic symptoms often emerge against a background of
more common symptoms such as depression and anxiety (Evins et al., 2005; Freeman
& Garety, 2003; Krabbendam & van Os, 2005; Wigman et al., 2012).
Another theory is often described as ‘the dopamine hypothesis’, which proposes that
psychosis and schizophrenia might be caused by over activity in the dopamine
neurotransmitter system in the mesolimbic system of the brain (Kapur & Mamo,
2003). The main evidence to support this theory is that effective drug treatment for
psychosis and schizophrenia regulates the dopaminergic neurotransmitter system.
However, a distinction must be made between the established pharmacological
action of antipsychotic drugs (which block dopamine receptors), and the hypothesis
that schizophrenia is caused by excessive activity of dopaminergic neurones, for
which the evidence is not clear-cut. For example, it could be that antipsychotic drugs
cause a general neurological suppression that reduces the intensity of symptoms
(Moncrieff, 2009).
Theories have also been put forward to explain how psychological factors may lead
to the development of psychotic symptoms. Psychological factors can be divided
into problems with basic cognitive functions, such as learning, attention, memory or
planning, and biases in emotional and reasoning processes. Problems in basic
cognitive functions are related to research in brain structure and function, while
problems with emotional and reasoning processes may be linked to social factors.
Both types of psychological factor have been implicated in the development of
symptoms of psychosis and schizophrenia (Garety et al., 2007; Garety et al., 2001;
Gray et al., 1991; Green, 1992; Hemsley, 1993). Hence studies of psychological factors
can provide a link between biological and environmental risk factors (van Os et al.,
2010).

On balance it is unlikely that any of these theories fully captures the complexity of
the potential gene-environment interaction that underpins the development of
psychosis and schizophrenia; see (van Os et al., 2010) for a detailed review of the
potential complexity of these interactions.
2.2 ASSESSMENT, ENGAGEMENT, CONSENT AND THE

THERAPEUTIC ALLIANCE
Assessment involves gathering information about current symptoms, the effects of
these symptoms on the individual (and their families and carers) and strategies the
person has developed to cope with them. Assessment provides an opportunity to
thoroughly examine the biological, psychological and social factors that may have
contributed to the onset of the illness, and also enquire about common coexisting
problems such as substance misuse, anxiety, depression and physical health
problems.
Assessments are carried out for a number of reasons primarily to establish a

diagnosis, as a means of screening (for example, for risk), to measure severity and
change and as the basis for a psychological formulation. Psychological formulations
provide an explanation of why a problem has occurred and what is maintaining it;
they also guide the intervention and predict potential difficulties that might arise.
The significant factors within the formulation will be underpinned by the theoretical
persuasion of the practitioner, including cognitive behavioural, systemic or
psychodynamic. A formulation is a hypothesis, based on the information that is
available at the time and will often be developed or change during the course of the
intervention. Although set in the context of a theoretical model, the formulation is
individualised based on the unique life experiences of each person. The individual
with psychosis or schizophrenia may not share professionals’ view of what the main
problem is. Seeking out and assisting with what the individual regards as the main
problem can provide a route towards establishing common ground, which may help
to establish trust and collaboration and allow collaborative care planning over time.
The development of a constructive therapeutic relationship is crucial to assessing

and understanding the nature of a person’s problems and provides the foundation of
any subsequent management plan. Engaging effectively with an individual with
psychosis or schizophrenia may require persistence, flexibility, reliability,
consistency and sensitivity to the individual’s perspective in order to establish trust.
Involving carers, relatives and friends of individuals with psychosis, and
acknowledging their views and needs, is also important in the process of assessment
and engagement, and in the long-term delivery of interventions (Kuipers &
Bebbington, 1990; Worthington et al., 2013).
At times people with acute psychosis may be intensely distressed, fearful, suspicious
and agitated or angry as psychotic symptoms can have a profound effect on a
person’s judgment and their capacity to understand their situation. They may
present a risk to themselves or others that justifies compulsory treatment or
detention. Issues of consent remain important throughout the care pathway and
professionals need to be fully aware of all appropriate legislation, particularly the
Mental Health Act (HMSO; Sartorius, 2002) and the Mental Capacity Act (HMSO).
All reasonable steps need to be taken to engage individuals in meaningful discussion
about issues relating to consent, and discussion with individuals should include
specific work around relapse signatures, crisis plans, advance statements and
advance decisions. The above statutory framework does provide for individuals
with schizophrenia to make a contemporaneous decision to refuse treatment, though
this could potentially be overruled by detention under the Mental Health Act.
In 2011-12, 48,631 individuals in England were compulsorily detained in hospital

under Mental Health Act provisions, showing a continuation of the increasing trend
in recent years (Care Quality Commission, 2012). There was also a 10% rise in the
number of inpatients made subject to community treatment orders (CTOs) to 4,220.
The CQC report identified concerns regarding inappropriate coercion in the system.
The awareness among individuals who have a psychotic disorder, their carers,
professionals and the general population that compulsory detention and treatment is
a possibility forms a key component in the mental health landscape, which is
variously seen as coercive, oppressive, enabling or protective. Therefore it is
essential that any individual detained under the Mental Health Act continues to be
involved in a collaborative approach to their difficulties. Seeking common objectives
is a vital part of this process and individuals subject to the provisions of the Mental
Health Act need the highest quality of care from the most experienced and trained
staff, including consultant psychiatrists.
2.3 LANGUAGE AND STIGMA

Although treatment for psychosis and schizophrenia has improved since the 1950s
and 1960s, some people with this diagnosis still encounter difficulties finding
employment and may feel excluded from society. In an editorial for the British
Medical Journal, Norman Sartorius claimed that ‘stigma remains the main obstacle to
a better life for the many hundreds of millions of people suffering from mental
disorders’ (Sartorius, 2002). In part because of media coverage of events associated
with psychosis and schizophrenia, people with the condition live with the stigma of
an illness often seen as dangerous and best dealt with away from the rest of society.
In this regard, research has shown that while the number of psychiatrically
unrelated homicides rose between 1957 and 1995, homicides by people sent for
psychiatric treatment did not, suggesting that the public fear of violence arising from
people with schizophrenia is misplaced (Taylor & Gunn, 1999).
Those with psychosis and schizophrenia may also feel stigmatised because of mental
health legislation, including compulsory treatment in the community, which may
exacerbate their feelings of exclusion. The side effects of the medication, such as
hypersalivation, involuntary movements, sedation and severe weight gain, and the
less than careful use of diagnostic labels, can all contribute to singling out people
with schizophrenia, marking them as different. In addition, people with this
condition may find that any physical health problems they have are not taken as
seriously by healthcare professionals.

In the view of many service users, clinical language is not always used in a helpful
way, and may contribute to the stigma of psychosis and schizophrenia. For example,
calling someone a ‘schizophrenic’ or a ‘psychotic’ gives the impression that the
person has been wholly taken over by an illness, such that no recognisable or
civilised person remains. Many non-psychiatric health workers and many employers
continue to approach people with psychotic disorders in this way. There is a move
away from using the word ‘schizophrenia’ for people with psychotic symptoms
because the label is so unhelpful, especially in early intervention in psychosis
services.
It is important that professionals are careful and considerate, but also clear and
thorough in their use of clinical language and in the explanations they provide, not
only to service users and carers but also to other healthcare professionals. Services
should also ensure that all clinicians are skilled in working with people from diverse
linguistic and ethnic backgrounds, and have a process by which they can assess
cultural influences and address cumulative inequalities through their routine clinical
practice (Bhui et al., 2007). Addressing organisational aspects of cultural competence
and capability is necessary alongside individual practice improvements.
Parents of people with psychosis and schizophrenia often feel to blame, either
because they believe that they have ‘passed on the genes’ causing schizophrenia, or
because they are ‘bad parents’. However, the families of people with schizophrenia
often play an essential part in the treatment and care of their relative, and with the
right support and help can positively contribute to promoting recovery. The caring
role can come at a high cost of depression and strain, and services need to remain
sensitive to the separate needs of carers (see Section 2.4).
2.4 ISSUES FOR FAMILIES, CARERS AND FRIENDS

This guideline uses the term ‘carer’ to apply to all people who provide or intends to
provide unpaid care or support for the person, including family members, friends
and advocates, although some family members may choose not to be carers.
Many people with psychosis and schizophrenia receive significant support from
carers and it is important to understand, therefore, that the caring role brings with it
many difficult challenges for which they may not be prepared. Carers may often be
important in the process of assessment and engagement in treatment and also in the
successful delivery of effective interventions and therapies for people with psychotic
disorders. As a result developing and sustaining supportive relationships with
carers may be instrumental for recovery from psychosis and schizophrenia.
Carers will need detailed information about psychosis and schizophrenia and, with
consent 1, will need guidance on their involvement in the person’s treatment and
1See http://www.carersandconfidentiality.org.uk for an interactive guide for professionals.

care. In such roles carers have rights and entitlements and these are described by the
NHS in England 2. Carers can be engaged in the care planning process by
practitioners drawing on good practice examples such as the ‘Triangle of Care’
(Kuipers & Bebbington, 1990; Worthington et al., 2013)
Caring for a person with psychosis or schizophrenia can be emotionally,

psychologically and financially challenging, therefore carers will need help and
support not only in their caring role but also for their own wellbeing because they
may experience grief, fear, distress and isolation, and these feelings can have a
significant impact on their quality of life. Without this support carers can feel
neglected by health and social care services in terms of their own health and support
needs and become frustrated by the lack of opportunities to contribute to the
development of the care plan for the person for whom they care.
2.5 TREATMENT AND MANAGEMENT OF PSYCHOSIS

AND SCHIZOPHRENIA IN THE NHS
2.5.1 Introduction
From the 1850s to the 1950s, the treatment and management of psychosis and
schizophrenia generally took place in large asylums where many people remained
confined for much of their lives. Subsequently, the development of the post-war
welfare state, which made benefits and housing more readily available in the
community, the introduction of antipsychotic drugs and increased concern with the
human rights of people with mental health problems have supported a government
policy of gradual closure of most asylums (Killaspy, 2006). Similar
deinstitutionalisation processes have taken place at varying rates in the USA and
most European countries, often aimed both at improving people’s quality of life and
reducing costs.
2.5.2 Pharmacological treatment

Today, within both hospital and community settings, antipsychotic drugs remain the
primary treatment for psychosis and schizophrenia. There is well-established
evidence for their efficacy in both the treatment of acute psychotic episodes and
relapse prevention over time (Horst et al., 2005). However, despite this, considerable
problems remain. A significant proportion of service users – up to 40%(Kelly et al.,
2008; Sacco et al., 2009) – have a poor response to conventional antipsychotic drugs
and continue to show moderate to severe psychotic symptoms (both positive and
negative).
In addition, conventional or typical antipsychotic agents (more recently called ‘first-

generation’ antipsychotics [FGAs]) are associated with a high incidence and broad
range of side effects including lethargy, sedation, weight gain and sexual
2http://www.nhs.uk/CarersDirect/guide/rights/Pages/carers-rights.aspx.

dysfunction. Movement disorders, such as parkinsonism, akathisia and dystonia
(often referred to as acute extrapyramidal side effects [EPS]), are common and can be
disabling and distressing. A serious long-term side effect is tardive dyskinesia,
which develops in around 20% of people receiving FGAs (Weinberger et al., 2008);
this is a late-onset EPS characterised by abnormal involuntary movements of the
lips, jaw, tongue and facial muscles, and sometimes the limbs and trunk. Although a
person who develops tardive dyskinesia is usually unaware of the movements, they
are clearly noticed by others, and the condition has long been recognised as a severe
social handicap (Williams et al., 2012b).
In response to the limited effectiveness and extensive side effects of FGAs,

considerable effort has gone into developing pharmacological treatments for
schizophrenia that are more effective and produce fewer or less disabling side
effects. The main advantage of these so-called second-generation (‘atypical’)
antipsychotics (SGAs) appears to be that they have a lower liability for acute EPS
and tardive dyskinesia. However, in practice this must be balanced against other
side effects, such as weight gain and other metabolic problems that may increase the
risk of type-2 diabetes and CVD (Lindenmayer et al., 2003; Mackin et al., 2007a;
Marder et al., 1996; Nasrallah, 2003; Nasrallah, 2008; Suvisaari et al., 2007). There
have been several recent suggestions that the distinction between FGAs and SGAs is
artificial (Kendall, 2011; Leucht et al., 2013).
Raised serum prolactin is also an important adverse effect of antipsychotic

medication, which can lead to problems such as menstrual abnormalities,
galactorrhea and sexual dysfunction, and in the longer term to reduced bone mineral
density (Haddad & Wieck, 2004; Meaney et al., 2004).
In people with schizophrenia who have not responded well to other antipsychotics,
only one antipsychotic drug, clozapine, has a specific licence for the treatment of this
group of people.
There is emerging evidence that some people can cope well in the long term without
antipsychotic medication (Harrow et al., 2012), and some suggestions that both
neurocognitive and social functioning may be improved without such medication
(Faber et al., 2012; Wunderink et al., 2013); in addition, there is preliminary evidence
that psychological interventions can be beneficial without antipsychotic medication
(Morrison et al., 2012b). Such considerations have led some to question the default
reliance on medication as first-line treatment for people with schizophrenia
(Morrison et al., 2012a). Nevertheless, it is widely accepted that antipsychotics
remain an essential component and not the mainstay of treatment (Kendall 2011).
2.5.3 Psychological and psychosocial interventions

Before the introduction of neuroleptic medication for schizophrenia in the 1950s and
1960s, analytical psychotherapies based on the work of Frieda Fromm-Reichmann
(1950) and Harry Stack Sullivan (1947) and others were widely practiced. The
concept of rehabilitation grew during this period influenced by the pioneering work

of Manfred Bleuler in the Bergholzi clinic in Zurich where patients were engaged in
meaningful vocational and occupational endeavour in the context of an ‘open door’
policy (Bleuler, 1978). In the early 1980s, the publication of the seminal ‘Chestnut
Lodge’ evaluation of exploratory and investigative psychotherapies (McGlashan,
1984) had a major impact: the trial demonstrated no impact of psychotherapy on the
core psychotic symptoms contributing to a decline in their use in routine practice
with neuroleptics taking their place as the mainstay of treatment.
However, as deinstitutionalisation gained ground in the 1970s, psychological and

social research into factors that might contribute to relapse in people with psychosis
or schizophrenia living in community settings, such as stressful life events and
communication difficulties in families (high ‘expressed emotion’), stimulated the
development of family intervention to prevent relapse (Leff et al., 1982; Lobban &
Barrowclough, 2009). Family intervention often included education for family
members about schizophrenia (sometimes called ‘psychoeducation’) and, in time,
research was conducted on the benefits of psychoeducation alone (Birchwood et al.,
1992).
Interest in psychological and broader psychosocial interventions for the treatment of

psychosis and schizophrenia was also precipitated in the 1980s by the increasing
recognition of the limitations, side effects and health risks associated with
antipsychotic medication and low rates of adherence (Akbarpour et al., 2010) and
growing evidence for the impact of cumulative neuroleptic exposure on cortical grey
matter loss (Baker et al., 2006).
Over the last decade, there has been a revolution in understanding the role that
ecological and psychological processes have on the risk for psychosis and on
resilience (Bloch et al., 2010). This includes, for example, the impact of urban
upbringing and residence in unstable, fragmented neighbourhoods (Chen et al.,
2013) and the impact that low self-esteem can have on the way in which individuals
with psychotic experience appraise its meaning.
Demand for psychological therapies in general has also grown, culminating in the
Department of Health’s Improving Access to Psychological Therapies (IAPT)
initiative; indeed, in the mental health strategy, No Health Without Mental Health
(Prince et al., 2007), funding has been made available to extend IAPT to those with
severe mental illness, particularly psychosis and schizophrenia.
Cognitive-developmental processes in psychosis

The familiar notion that the onset of psychosis coincides with the ‘first psychotic
episode’ is now understood to be something of a misnomer; it is, in reality, the ‘end
of the beginning’. With few exceptions, the formal onset of psychosis is preceded by
many months of untreated psychosis and before that, many years of changes
stretching back into late childhood. Important prospective studies, particularly the
‘Dunedin Study’(Dalack & Meador-Woodruff, 1999), have shown that subtle
psychotic-like experiences at age 11 strongly predict the later emergence of

psychosis; however many individuals manage to escape this outcome. Population
studies such as the NEMESIS project (de Leon et al., 2005) and the UK AESOP study
(Chen et al., 2013) have shown that a number of ‘environmental’ factors predict those
who are more likely to show persistence and worsening of symptoms, including:
cannabis exposure in adolescence, social deprivation, absence of a parent and the
experience of childhood abuse or neglect. Affective dysregulation has been shown to
be a dimension that is both highly comorbid with psychosis (now argued to be a
dimension of psychosis) and a strong feature in its early development (Evins et al.,
2005); the presence of affective dysfunction in adolescence, particularly depression
and social anxiety, has been shown to be a predictor of transition from psychotic
experience to psychotic disorder (Bloch et al., 2010).
Social disability is one of the hallmarks of psychosis and those with adolescent onset
tend to fare worse in this regard. Prospective studies of social disability and recovery
have shown that early functional and vocational recovery, rather than symptoms of
psychosis, play a pivotal role in preventing the development of chronic negative
symptoms and disability, underlining the need for interventions that specifically
address early psychosocial recovery (Fatemi et al., 2005).
These cognitive-developmental processes have informed influential cognitive

models of psychosis (Gallagher et al., 2007) and specific symptoms of psychosis such
as auditory hallucinations (Gelkopf et al., 2012; George et al., 2008) and affective
processes (George et al., 2000). These models have informed wider foci of
interventions in psychosis in addition to psychotic symptoms, embracing the family,
developmental trauma and their adult sequelae, affective dysfunction, substance
misuse and peer social engagement.
Aims of psychological and psychosocial interventions

The aims of psychological and psychosocial interventions in psychosis and
schizophrenia are therefore numerous. These should include interventions to
improve symptoms but also those that address vulnerability, which are embedded in
developmental processes. The aims, therefore, include: reduction of distress
associated with psychosis symptoms (Hartman et al., 1991); promoting social and
educational recovery; reducing depression and social anxiety (Hong et al., 2011); and
relapse prevention. Reducing vulnerability and promoting resilience will require
reducing cannabis misuse, promoting social stability and family support, and
dealing with the sequelae of abuse and neglect including attachment formation.
2.5.4 Management of at risk mental states and early psychotic

symptoms
Reliable and valid criteria are now available to identify help-seeking individuals in
diverse settings who are at high risk of imminently developing schizophrenia and
related psychoses. Yung and colleagues (Yung et al., 1996) developed operational
criteria to identify three subgroups possessing an at risk mental state for psychosis.
Two subgroups specify state risk factors, defined by the presence of either transient
psychotic symptoms, also called brief limited intermittent psychotic symptoms, or

attenuated (subclinical) psychotic symptoms. The other subgroup comprises trait-
plus-state risk factors, operationally defined by the presence of diminished
functioning plus either a first-degree relative with a history of psychosis or a pre-
existing schizotypal personality disorder. All subgroups are within a specified age
range known to be at greatest risk for the onset of psychosis.
Effective interventions to prevent or delay transition to psychosis are needed

because of the significant personal, social and financial costs associated with it. To
date there have been six randomised controlled trials (RCTs) that have reported
outcomes associated with antipsychotic medication, omega-3 polyunsaturated fatty
acids and/or psychological interventions, each using similar operational definitions
of at risk mental states. These studies have been conducted in Australia (McGorry et
al., 2002; Yung et al., 2011), North America (Addington et al., 2011; McGlashan et al.,
2006); the UK (Morrison et al., 2007; Morrison et al., 2004) and Austria (Amminger et
al., 2010).
It is generally agreed that research regarding interventions for at risk mental states
and subthreshold psychotic experiences is in a state of clinical equipoise. Existing
recommendations promote a clinical staging approach that utilises benign
interventions (such as monitoring mental states, case management, social support
and psychosocial interventions) before considering those with more significant side
effects, such as antipsychotic medication, or restrictive approaches involving
hospitalisation (International Early Psychosis Association Writing Group, 2005;
McGorry et al., 2006). However, due to local resources and service configurations,
clinicians’ attitudes and awareness of such recommendations, current clinical
practice is likely to be highly variable, which is evident in the recent large
international naturalistic cohort studies (Cannon et al., 2008; Ruhrmann et al., 2010).
2.5.5 Service-level interventions

Service-level interventions for people with psychosis and schizophrenia are
delivered both in hospital and in community settings. The ‘balanced care’ model of
mental health service provision (Thornicroft & Tansella, 2012) emphasises the
importance of achieving an equilibrium among all service components including
outpatient services and community mental health teams, acute inpatient services,
community residential care and services for supporting employment.
Despite the policy of shifting care to the community, expenditure on inpatient care
remains substantial: secure units, community mental health teams and acute wards
are the top three sources of mental health expenditure in the NHS (Nayor & Bell,
2010). As the large asylums closed, government policy promoted the opening of
acute psychiatric units within general hospitals. Some such units remain, but
recently the separation of mental health provider trusts from physical health
services, together with disappointment with the extent to which mental healthcare in
the general hospital has reduced stigma, has resulted in a trend towards small
freestanding mental health inpatient units, usually within or close to the catchment
areas they serve (Totman et al., 2010). Both service users and clinicians have argued

that general acute admission wards are often unsafe environments with limited
provision of therapeutic interventions and activities (Holloway & Lloyd, 2011). In
response, there has been a series of initiatives aimed at improving the quality and
effectiveness of inpatient care, including the Accreditation for Acute Inpatient
Mental Health Services (AIMS) programme initiated by the Royal College of
Psychiatrists (Cresswell & Lelliott, 2009) and STAR WARDS (Simpson & Janner,
2010).
Beyond the acute admission ward, there has been interest for many decades in
whether residential crisis houses outside hospital can provide effective and
acceptable alternatives to hospital admission for some people who have severe
mental illness. Service users and voluntary sector organisations have strongly
advocated them. They are available in a minority of trusts and are often closely
connected to crisis resolution and home treatment teams (Johnson et al., 2010). While
numbers of acute beds have fallen, secure bed use for longer term admission of
people deemed too dangerous for local psychiatric units has increased (Walker et al.,
2012). This trend, together with a rise in supported housing and in detentions under
the Mental Health Act, has led some to argue that a reinstitutionalisation process is
in progress (Priebe et al., 2005).
The lynchpin of community mental healthcare for people with a psychotic disorder
in the past 2 decades has been the multidisciplinary community mental health team,
providing assessment and long-term follow-up. Mandated by the NHS Plan (2000), a
strikingly extensive national initiative has been the introduction in every catchment
area in England of three types of specialist community mental health teams: (1) crisis
resolution and home treatment teams provide urgent assessment when hospital
admission is contemplated and, where feasible, offer intensive home treatment as an
alternative (Johnson et al., 2008); (2) assertive outreach (assertive community
treatment) teams work intensively with people who are most difficult to engage
(Wright et al., 2003); and (3) early intervention in psychosis services seek to reduce
treatment delays at the onset of psychosis and to promote recovery and reduce
relapse following a first episode of psychosis (Lester et al., 2009a). With a new
government in 2010 and a shift towards focusing on outcomes rather than requiring
certain service configurations, these new team types are no longer mandatory, but
they remain important components of service systems in most local areas. In some
regions, generic community mental health teams are now giving way to further
types of specialist service, including primary care liaison teams and specialist teams
for psychosis. In recent innovations, there has been a further focus on the
development of integrated pathways through services: for example, in some
catchment areas integrated acute care pathways closely integrate inpatient wards,
crisis teams, crisis houses and acute day services, with a single management
structure and sometimes staff rotation between services. Rehabilitation services,
often consisting of inpatient, residential and community team components, are a
longstanding resource for people with psychosis and schizophrenia in many areas,
focusing on people with treatment-resistant symptoms and severe difficulties in
functioning (Killaspy et al., 2013).

A great variety of services aim to meet the social needs of people with psychosis and
schizophrenia. Recent emphasis has been on developing services that support people
in achieving their own self-defined recovery goals. As the National Institute for
Mental Health in England (NIMHE) stated: ‘Recovery is what people experience
themselves as they become empowered to manage their lives in a manner that
allows them to achieve a fulfilling, meaningful life and a contributing positive sense
of belonging in their communities’ (National Institute for Mental Health in England,
2005). The social disadvantages experienced by people with severe mental illness,
including stigma, social exclusion and poverty, are still great, therefore high levels of
need in domains such as accommodation, work, occupational, educational and social
activities, and social support remain unaddressed (Thornicroft et al., 2004). A
complex range of supported accommodation, varying in quality, support level and
approach, is delivered primarily by the voluntary and private sectors (Macpherson
et al., 2012). Employment rates among people with severe mental illness are notably
low in the UK, and a range of services, including individual placement and support
schemes (Rinaldi et al., 2010) and social firms (which seek to create jobs for people
who are disadvantaged in the labour market) have sought to address this. Social
support and non-vocational activities have traditionally been the province of local
authority day centres. These have sometimes been criticised as excessively
institutional, and have been supplemented or replaced by a wider range of
initiatives aimed at improving access to meaningful activities, enhancing personal
relationships, reducing stigma and discrimination, and lessening the negative effects
of social isolation. Many such innovative services are provided by the voluntary
sector, but relatively little evidence on activities and outcomes is available as yet. See
Section 2.5.6 for further discussion about employment for people with psychosis and
schizophrenia.
2.5.6 Employment
When people have a job that gives them purpose, structure and a valued role in
society this impacts positively on their self-esteem, community inclusion and
opportunities (Ross, 2008) as well as having a financial reward, although there are
many positive benefits to unpaid work. Conversely, unemployment limits life
chances and has a detrimental impact on physical health, social networks and choice
(Advisory Conciliation and Arbitration Service, 2009).
Rates of unemployment for people with severe mental illness are approximately six
to seven times higher than people with no mental disorder (Organisation for
Economic Co-operation and Development, 2011). Different studies put the
employment rate of people with severe mental illness in a range of between 15%
(Evans & Repper, 2000) to 20% (Schneider et al., 2007), and they are the largest group
claiming incapacity benefit (Ross, 2008).
For people with a severe mental illness, the best predictor for a positive outcome
towards an employment goal is the service user wanting to have a work role (Ross,
2008) and a work history (Michon et al., 2005), rather than the diagnosis or

symptoms. Having unmet needs and not receiving incapacity benefit or income
support was associated with wanting to work full-time (as opposed to part time)
rather than self-esteem, quality of life, severity of symptoms or level of functioning
(Rice et al., 2009).
The stress-vulnerability model can lead to the view that work could be detrimental
to people with psychosis and schizophrenia because it could be stressful (Zubin &
Spring, 1977). But having little structure or role in society, which can lead to social
isolation and poverty, are widely recognised as stressors (Marrone & Golowka, 1999)
and contributors to poor physical and mental health (Boardman et al., 2003). If health
and social care professionals assume that service users do not want to work and
suggest that work may be an unreasonable aspiration or too stressful, this will limit
the views of the service user. Low expectations of mental health staff can be a major
barrier to service users finding employment (Office of the Deputy Prime Minister,
2004). There is evidence that up to 97.5% of service users may want some type of
work role, be that volunteering or paid employment, but when asked if they had any
help with seeking work, 53% had not received any support with this goal (Seebohm
& Secker, 2005).
Stigma and discrimination is experienced by people with psychosis and

schizophrenia from employers, with 75% of employers stating that it would be
difficult to employ a person with a psychotic disorder (Office of the Deputy Prime
Minister, 2004). Some employers believe that workers with mental health problems
cannot be trusted and cannot work with the public and that work would be negative
to their mental health. Larger employers are more likely to employ people with
psychosis and schizophrenia, perhaps because they have wider support structures
(Biggs et al., 2010). Service users identified the attitude of employers as the biggest
barrier to work (Seebohm & Secker, 2005). However, the attitude of employment
agencies has improved and they were able to identify the advantages of employment
for service users (Biggs et al., 2010).
Other barriers to employment identified by service users with mental health

problems are the benefits system and having a lack of work experience, skills and
qualifications (Seebohm & Secker, 2005). One key determinant that can limit
employment outcomes is the level of educational attainment. Experiencing
disruption to education as a direct result of mental health problems can impact on
access to the labour market and can make it difficult to attain and sustain a work role
(Organisation for Economic Co-operation and Development, 2011; Schneider et al.,
2009). Even for healthy young people there is evidence for long-term negative effects
on their work prospects when, having completed their education, they are unable to
access the labour market during a recession; this can lead to subsequent anxiety
about job security because past unemployment will influence future expectations
and limit lifetime earnings (Bell & Blanchflower, 2011). Therefore, when a young
person’s future is compounded further by poor mental health, they require
exceptional support and guidance to achieve their occupational aspirations and

mental health workers need to be active in challenging the barriers that may be
inherent within the system for service users to achieve their full potential.
2.5.7 Inequalities
The Equality Act identifies the following characteristics that require protection
against discrimination in relation to service provision: age, race, religion or belief,
gender, sexual orientation, transgender identity, disability and pregnancy and
maternity. Marriage or civil partnership relates only to employment. It is important
for service providers and mental health workers to be aware of the different needs
and outcomes for people with protected characteristics, and how these may affect
the way that services and interventions are designed, accessed, delivered and
evaluated. As a result of this information, services need to take equality into account
in working with individuals or population groups, so that they can demonstrate that
people within these characteristics are not disadvantaged in their care and
subsequent outcomes and address health inequalities.
Many of the protected characteristics, such as race, age, perinatal mental health and
gender, have been covered widely in the literature in relation to psychosis and
schizophrenia. The evidence base is non-existent in relation to the population that
have protected characteristics relating to sexual orientation, gender reassignment
and disability. However, current evidence demonstrates lesbian, gay and bisexual
people have a higher prevalence of self-harm, suicidal ideation, substance misuse
(Hunt & Fish, 2008) (Stonewall, 2012) and are frequent victims of bullying and hate
crime from family members and within society (Dick, 2008) and subsequent
psychological trauma (Herek et al., 1999).
2.5.8 Primary and secondary care interface

The last decade has seen much change in how the care of people with psychosis and
schizophrenia living in the community is organised between primary and secondary
care. Not only has secondary care provision undergone major alteration but there
have also been significant changes in primary care provision. A recent 12-month
investigation of 1,150 primary care records of people with severe mental illness—the
most common diagnoses being schizophrenia (56%) and bipolar disorder (37%)—
from 64 practices in England (Reilly et al., 2012) found that per annum about two
thirds were seen by a combination of primary and specialist services and a third
were seen just in primary care. These findings superficially appeared similar to
findings from the largest previous survey (Kendrick et al., 1994). However this new
study (Reilly et al., 2012) revealed a marked reduction in this population’s annual
general practitioner (GP) consultation rates averaging only 3 (range 2–6) per annum,
far lower than the rates of 13 to 14 per annum reported in the mid-1990s (Nazareth &
King, 1992), and only slightly higher than the annual consultation rate of the general
population at 2.8 (range 2.5–3.2) in 2008 (Hippisley-Cox & Vinogradova, 2009).
Moreover practice nurses, key providers of cardiovascular risk screening and health
education in primary care, consulted with this population on average only once a
year compared with the general practice population rate of 1.8 consultations per
year; nor was health education a common feature of these consultations, the authors

concluding that practice nurses appear to be an underutilised resource (Reilly et al.,
2012). This diminution in contact with a primary care practitioner is perhaps
surprising given that in 2006 the Quality and Outcomes Framework (NHS
Employers and British Medical Association 2011/12) instituted a pay for
performance scheme designed to encourage health promotion and disease
management programmes, paying primary care to measure four physical health
indicators for people with severe mental illness on the primary care mental illness
register: BMI (MH12), blood pressure (MH13), total to HDL cholesterol ratio (MH14)
and blood glucose (MH15).
Patients view primary care as providing an important coordinating role for their
mental and physical healthcare; they particularly value a stable continuity of doctor–
patient relationship in primary care (Lester et al., 2005). In contrast GPs report
feeling that the holistic care of people with severe mental illness is beyond their
remit (Lester et al., 2005); some may hold negative opinions about providing care for
this population (Curtis et al., 2012; Lawrie et al., 1998); and the majority regard
themselves as simply involved in the monitoring and treatment of physical illness
and prescribing for mental illness (Bindman et al., 1997; Kendrick et al., 1994).
Detection and referral of psychosis

The pathway to effective assessment and treatment for someone with a newly
presenting psychotic illness is an important aspect of the primary–secondary
interface. Rarity of presentation of psychotic disorders in primary care can impede
early detection, highlighted by a Swiss study that found that GPs suspect an
emerging psychosis in only 1.4 patients per year (Simon et al., 2005). Yet GP
involvement is linked with fewer legal detentions and can reduce distress (Burnett et
al., 1999; Cole et al., 1995). However, few GPs receive postgraduate mental health
training, and even when they do a well-powered study of a GP educational
intervention about early presentations of psychosis failed to reduce treatment delay,
although the training may have facilitated access to specialist early intervention
teams (Lester et al., 2009b). When asked, GPs prefer greater collaboration with
specialist services and low-threshold referral services rather than educational
programmes (Simon et al., 2005).
Coordination of physical healthcare

The other major interface issue concerns the management of physical health. A
Scottish primary care study confirmed the high rates of multiple comorbid physical
health problems experienced by people with schizophrenia, and that the likelihood
of comorbidity was almost doubled for those living in the most deprived areas
(Langan et al., 2013). There is evidence from studies in the general population that
the extent of comorbidity is greater in younger age groups, even though there is
increasing morbidity with age (van den Akker et al., 1998). This is particularly
pertinent for people experiencing schizophrenia, where young onset and social
disadvantage are both likely.

Cardiovascular disease (CVD) is the single commonest cause of premature mortality
in people with psychosis and schizophrenia and yet, despite numerous published
screening recommendations in this guideline and other reports (Buckley et al., 2005;
Mackin et al., 2007b; Morrato et al., 2009; Nasrallah et al., 2006), there continues to be
systematic under-recognition and under-treatment in primary care (Smith et al.,
2013). Recognition and treatment of CVD risk was one of the themes investigated by
the recent National Audit of Schizophrenia (Royal College of Psychiatrists, 2012)
using standards derived from the 2009 guideline (NICE, 2009d). In the largest audit
of its kind yet undertaken, 94% of the trusts and health boards across England and
Wales took part, returning data between February and June 2011 on 5,091 patients
with an average age of 45 years. This case record audit reviewed the care of people
with a diagnosis of either schizophrenia or schizoaffective disorder in contact with
community-based mental health services in the previous 12 months. Only 29% had
record of a comprehensive assessment of cardiovascular risk, including weight (or
BMI), smoking status, blood glucose, blood lipid levels and blood pressure; 43%
appeared not to have been weighed and 52% had information about family history
of CVD, diabetes, hypertension or hyperlipidaemia during the previous 12 months.
Of those with an established comorbidity of either CVD or diabetes mellitus, fewer
than half had record of a comprehensive assessment of cardiovascular risk. Even
where monitoring had identified a problem, an intervention did not necessarily
occur – for instance only 20.1% of those identified to have a lipid abnormality appear
to have been offered an intervention.
Perhaps because poor physical health may take several years to fully develop in
people with psychosis and schizophrenia, there has been a tendency for most
guidance and recommendations to focus on treating the endpoints of disease. Yet
modifiable cardiovascular risk appears within weeks of commencing treatment
(Foley & Morley, 2011). New models are, however, emerging. For instance, the
potential for nurse-led approaches to cardiovascular risk screening has attracted
interest. A recent study designed to complement the configuration of UK primary
and secondary care services placed a general nurse, experienced in cardiovascular
risk assessment but without previous mental health experience, within four
community mental health teams; the nurse-led intervention was superior, resulting
in an absolute increase of approximately 30% more people with serious mental
illness receiving screening for each CVD risk factor than in control arm of the study
(Osborn et al., 2010a). Another model, recently introduced in New South Wales is
encouraging a systematic approach by specialist services for people with first
episode psychosis based on an agreed clinical algorithm focusing on key
cardiovascular risks – notably weight gain, smoking, lipid and glucose
abnormalities, hypertension, awareness of family history of CVD or diabetes (Curtis
et al., 2012). This resource has recently been adapted for use in the UK by the Royal
College of General Practitioners and the Royal College of Psychiatrists as part of the
National Audit of Schizophrenia initiative; the Positive Cardiometabolic Health
Resource (Lester UK adaptation, 2012) encourages a collaborative framework
between primary and specialist care for dealing with the cardiometabolic risks
linked to prescribing antipsychotic medicines.

While such examples of innovation and collaboration between professionals from
primary and specialist care are encouraging, there remains little systematic
evaluation of ways to better address multiple physical health morbidities in people
with psychosis and schizophrenia.
2.6 ECONOMIC COST

Schizophrenia is one of the main contributors to global disease burden (Collins et al.,
2011), having a significant impact on individuals and placing heavy responsibility on
their carers, as well as potentially large demands on the healthcare system. In the
most recent ‘Global Burden of Disease’ analysis by Murray and colleagues (2012)
schizophrenia appeared among the top 20 causes of disability in many regions and
was ranked as the 16th leading cause of disability among all diseases worldwide.
When the burden of premature mortality and non-fatal health outcomes were
combined and expressed in disability adjusted life years (DALYs), schizophrenia
was the 43rd leading cause of worldwide burden among all diseases and from 1990
to 2010 there was a 43.6% increase in DALYs attributable to schizophrenia
worldwide. Similarly, in the UK sub-analysis of the ‘Global Burden of Disease’ study
Murray and colleagues (2013) found schizophrenia to be one of the leading causes of
years lived with disability (YLDs) with approximately 15% increase in YLDs and
14% increase in DALYs from 1990 to 2010.
In England schizophrenia is estimated to cost £7.9 billion (in 2011/2012 prices)

(Mangalore & Knapp, 2007). Of this, roughly £2.4 billion (about 30% of the total cost)
comprise direct costs of treatment and care falling on the public purse, while the
remaining £5.6 billion (70% of the total cost) constitute indirect costs to society. The
cost of lost productivity of people with schizophrenia owing to unemployment,
absence from work and premature mortality reach £4.0 billion, while the cost of lost
productivity of carers is £38.0 million. The cost of informal care and private
expenditures borne by families, account for approximately £729.4 million. In
addition, £1.2 million of the total cost can be attributed to criminal justice system
services, £676.0 million to benefit payments and another £16.6 million to the
administration of these payments. Based on the above estimates, the average annual
cost of a person with schizophrenia in England is approximately £65,000.
Davies and Drummond (1994) estimated that the lifetime total direct and indirect
costs of a person with schizophrenia ranged from £8,000 (for a person with a single
episode of schizophrenia) to £535,000 (for a person with multiple episodes lasting
more than 2.5 years, requiring long-term care either in hospital or intensive
community programmes) in 1990/1991 prices. Guest and Cookson (1999) estimated
the average costs of a newly diagnosed person with schizophrenia at around
£115,000 over the first 5 years following diagnosis, or approximately £23,000
annually (1997 prices). Of these, 49% were indirect costs owing to lost productivity.
Schizophrenia has been shown to place a substantial economic burden to the

healthcare system and society worldwide: Wu and colleagues (2005) reported a total

cost of schizophrenia in the US of US$62.7 billion (2002 prices). More than 50% of
this cost was attributed to productivity losses, caused by unemployment, reduced
workplace productivity, premature mortality from suicide and family caregiving;
another 36% was associated with direct healthcare service use and the remaining
12% was incurred by other non-healthcare services. In Canada, Goeree and
colleagues (2005) estimated the total cost of schizophrenia at approximately CA$2.02
billion (2002 prices). Again, productivity losses were by far the main component of
this cost (70% of the total cost). In Australia, the total societal cost associated with
schizophrenia reached AU$1.44 billion in 1997/1998 prices, with roughly 60%
relating to indirect costs (Carr et al., 2003). Finally, several national studies
conducted in Europe in the 1990s showed that schizophrenia was associated with
significant and long-lasting health, social and financial implications, not only for
people with schizophrenia but also for their families, other caregivers and the wider
society (Knapp et al., 2004).
The use of hospital inpatient care by people with psychosis and schizophrenia is
substantial. In the financial year 2011–2012, 29,172 admissions were reported for
schizophrenia and related disorders in England, resulting in over 2.8 million
inpatient bed days. Moreover, there were approximately 56,000 outpatient
attendances and 2,700 teleconsultations related to the management of schizophrenia
and other psychotic disorders (The Health and Social Care Information Centre,
2012). Inpatient care is by far the most costly healthcare component in the overall
treatment of schizophrenia. Kavanagh and colleagues (1995) found that care in short-
or long-stay psychiatric hospitals accounted for 51% of the total public expenditure
on care for people with schizophrenia. Lang and colleagues (1997) reported that
provision of inpatient care for people with schizophrenia amounted to 59% of the
total cost of health and social care for this population. Similarly Knapp and
colleagues (2002) suggested that inpatient care accounted for 56.5% of the total
treatment and care costs of schizophrenia, compared with 2.5% for outpatient care
and 14.7% for day care. Unemployment is a considerable burden for people with
schizophrenia. A rate of employment among people with schizophrenia is reported
to be between 15 (Evans & Repper, 2000) and 20% (Schneider et al., 2007) in the UK.
Stigmatisation is one of the main barriers to employment for this population.
Generally the rates of employment are higher for newly diagnosed people compared
with those with established schizophrenia; however, the majority of people
presenting to services for the first time are already unemployed (Marwaha &
Johnson, 2004). According to Guest and Cookson (1999), between 15 and 30% of
people with schizophrenia are unable to work at diagnosis, rising to 67% following a
second episode. Overall, the estimates of total indirect costs of people with
schizophrenia in the UK range from £412 million for newly diagnosed people over
the first 5 years following diagnosis (Guest & Cookson, 1999) to £1.7 billion annually
for people with chronic schizophrenia (Davies & Drummond, 1994).
Family members and friends often provide care and support to those with
schizophrenia, which places significant burdens on them that impact upon their
health, leisure time, employment and financial status. Guest and Cookson (1999)

estimated that, in the UK, 1.2 to 2.5% of carers gave up work to care for dependants
with schizophrenia.
Measuring the total cost of informal care provided by family members and friends is
difficult but it is important to highlight that it is a significant amount. Data on costs
of informal care for people with schizophrenia are not available. Based on figures
provided by the Office for National Statistics, the Sainsbury Centre for Mental
Health (2003) estimated that in 2002/2003 the aggregate value of informal care
provided by family members and friends in the UK to those with mental health
problems was £3.9 billion.
It is therefore evident that efficient use of available healthcare resources is required

to maximise the health benefit for people with schizophrenia and, at the same time,
reduce the emotional distress and financial implications to society.

3 METHODS USED TO DEVELOP
THIS GUIDELINE
3.1 OVERVIEW
The development of this guideline followed The Guidelines Manual (NICE, 2012b). A
team of health care professionals, lay representatives and technical experts known as
the Guideline Development Group (GDG), with support from the NCCMH staff,
undertook the development of a person-centred, evidence-based guideline. There are
seven basic steps in the process of developing a guideline:
1. Define the scope, which lays out exactly what will be included (and
excluded) in the guidance.
2. Define review questions that cover all areas specified in the scope.
3. Develop a review protocol for the systematic review, specifying the search
strategy and method of evidence synthesis for each review question.
4. Synthesise data retrieved, guided by the review protocols.
5. Produce evidence profiles and summaries using the Grading of
Recommendations Assessment, Development and Evaluation (GRADE)
approach.
6. Consider the implications of the research findings for clinical practice and
reach consensus decisions on areas where evidence is not found.
7. Answer review questions with evidence-based recommendations for
clinical practice.
The clinical practice recommendations made by the GDG are therefore derived from
the most up-to-date and robust evidence for the clinical and cost effectiveness of the
interventions and services used in the treatment and management of people with
psychosis and schizophrenia in adults. Where evidence was not found or was
inconclusive, the GDG discussed and attempted to reach consensus on what should
be recommended, factoring in any relevant issues. In addition, to ensure a service
user and carer focus, the concerns of service users and carers regarding health and
social care have been highlighted and addressed by recommendations agreed by the
whole GDG.
3.2 THE SCOPE

Topics are referred by the Secretary of State and the letter of referral defines the
remit, which defines the main areas to be covered (see The Guidelines Manual (NICE,
2012b) for further information). The NCCMH developed a scope for the guideline
based on the remit (see Appendix 1). The purpose of the scope is to:
• provide an overview of what the guideline will include and exclude

• identify the key aspects of care that must be included

• set the boundaries of the development work and provide a clear
framework to enable work to stay within the priorities agreed by NICE
and the National Collaborating Centre, and the remit from the
Department of Health/Welsh Assembly Government
• inform the development of the review questions and search strategy
• inform professionals and the public about expected content of the
guideline
• Keep the guideline to a reasonable size to ensure that its development
can be carried out within the allocated period.
An initial draft of the scope was sent to registered stakeholders who had agreed to
attend a scoping workshop. The workshop was used to:
• obtain feedback on the selected key clinical issues

• identify which population subgroups should be specified (if any)
• seek views on the composition of the GDG
• Encourage applications for GDG membership.
The draft scope was subject to consultation with registered stakeholders over a 4-
week period. During the consultation period, the scope was posted on the NICE
website (www.nice.org.uk). Comments were invited from stakeholder organisations
The NCCMH and NICE reviewed the scope in light of comments received, and the
revised scope was signed off by NICE.
3.3 THE GUIDELINE DEVELOPMENT GROUP

During the consultation phase, members of the GDG were appointed by an open
recruitment process. GDG membership consisted of: professionals in psychiatry,
clinical psychology, nursing, social work, and general practice; academic experts in
psychiatry and psychology; and service users, carers and representatives from
service user and carer organisations. The guideline development process was
supported by staff from the NCCMH, who undertook the clinical and health
economic literature searches, reviewed and presented the evidence to the GDG,
managed the process, and contributed to drafting the guideline.
3.3.1 Guideline Development Group meetings

Eleven GDG meetings were held between Tuesday 28 February 2012 and Tuesday 15
October 2013. During each day-long GDG meeting, in a plenary session, review
questions and clinical and economic evidence were reviewed and assessed, and
recommendations formulated. At each meeting, all GDG members declared any
potential conflicts of interest (see Appendix 2), and service user and carer concerns
were routinely discussed as a standing agenda item.
3.3.2 Service users and carers

Individuals with direct experience of services gave an integral service-user and carer
focus to the GDG and the guideline. The GDG included two service users and a carer
representative of a national service user group. They contributed as full GDG

members to writing the review questions, providing advice on outcomes most
relevant to service users and carers, helping to ensure that the evidence addressed
their views and preferences, highlighting sensitive issues and terminology relevant
to the guideline, and bringing service user research to the attention of the GDG. In
drafting the guideline, there was regular communication with the NCCMH team to
develop the chapter on carer experience and they contributed to writing the
guideline’s introduction and identified recommendations from the service user and
carer perspective.
3.3.3 Special advisors

Special advisors, who had specific expertise in one or more aspects of treatment and
management relevant to the guideline, assisted the GDG, commenting on specific
aspects of the developing guideline and making presentations to the GDG.
Appendix 4a lists those who agreed to act as special advisors.
3.3.4 National and international experts

National and international experts in the area under review were identified through
the literature search and through the experience of the GDG members. These experts
were contacted to identify unpublished or soon-to-be published studies, to ensure
that up-to-date evidence was included in the development of the guideline. They
informed the GDG about completed trials at the pre-publication stage, systematic
reviews in the process of being published, studies relating to the cost effectiveness of
treatment and trial data if the GDG could be provided with full access to the
complete trial report. Appendix 5 lists researchers who were contacted.
3.4 REVIEW QUESTIONS

Review (clinical) questions were used to guide the identification and interrogation of
the evidence base relevant to the topic of the guideline. Before the first GDG
meeting, draft review questions were prepared by NCCMH staff based on the scope
(and an overview of existing guidelines), and discussed with the guideline Chair.
The draft review questions were then discussed by the GDG at the first few meetings
and amended as necessary. Where appropriate, the questions were refined once the
evidence had been searched and, where necessary, sub-questions were generated.
The final list of review questions and their protocols can be found in Appendix 6.
For questions about interventions, the PICO (Population, Intervention, Comparison

and Outcome) framework was used to structure each question (see Table 1).

Table 1: Features of a well-formulated question on the effectiveness of an
intervention – PICO
Population: Which population of service users are we interested in? How can they be
best described? Are there subgroups that need to be considered?
Intervention: Which intervention, treatment or approach should be used?
Comparison: What is/are the main alternative/s to compare with the intervention?
Outcome: What is really important for the service user? Which outcomes should be
considered: intermediate or short-term measures; mortality; morbidity
and treatment complications; rates of relapse; late morbidity and
readmission; return to work, physical and social functioning and other
measures such as quality of life; general health status?
In some situations, the prognosis of a particular condition is of fundamental

importance, over and above its general significance in relation to specific
interventions. Areas where this is particularly likely to occur relate to assessment of
risk, for example in terms of behaviour modification or screening and early
intervention. In addition, review questions related to issues of service delivery are
occasionally specified in the remit from the Department of Health/Welsh Assembly
Government. In these cases, appropriate review questions were developed to be
clear and concise.
To help facilitate the literature review, a note was made of the best study design type
to answer each question. There are four main types of review question of relevance
to NICE guidelines. These are listed in Table 2. For each type of question, the best
primary study design varies, where ‘best’ is interpreted as ‘least likely to give
misleading answers to the question’.
However, in all cases, a well-conducted systematic review (of the appropriate type of
primary study) is likely to always yield a better answer than a single study.
For reviews of interventions, if no existing systematic reviews address the review

question, then in the first instance only RCTs will usually be included. The range of
included studies will be expanded to controlled before-after studies and interrupted
time-series if the RCT evidence is inadequate to address the review question.

Table 2: Best study design to answer each type of question
Type of question Best primary study design
Effectiveness or other impact of an Randomised controlled trial (RCT); other studies that
intervention may be considered in the absence of RCTs are the
following: internally/externally controlled before and
after trial, interrupted time-series
Accuracy of information (for example, Comparing the information against a valid gold
risk factor, test, prediction rule) standard in an RCT or inception cohort study
Rates (of disease, service user Prospective cohort, registry, cross-sectional study
experience, rare side effects)
Experience of care Qualitative research (for example, grounded theory,
ethnographic research)
3.5 CLINICAL REVIEW METHODS

The aim of the clinical literature review was to systematically identify and synthesise
relevant evidence from the literature in order to answer the specific review questions
developed by the GDG. Thus, clinical practice recommendations are evidence-based,
where possible and, if evidence is not available, informal consensus methods are
used to try and reach general agreement between GDG members (see Section 3.5.6)
and the need for future research is specified.
3.5.1 The search process

Scoping searches
A broad preliminary search of the literature was undertaken in August 2011 to
obtain an overview of the issues likely to be covered by the scope, and to help define
key areas. Searches were restricted to clinical guidelines, Health Technology
Assessment (HTA) reports, key systematic reviews and RCTs. A list of databases and
websites searched can be found in Appendix 13.
Systematic literature searches

After the scope was finalised, a systematic search strategy was developed to locate as
much relevant evidence as possible. The balance between sensitivity (the power to
identify all studies on a particular topic) and specificity (the ability to exclude
irrelevant studies from the results) was carefully considered, and a decision made to
utilise a broad approach to searching to maximise retrieval of evidence to all parts of
the guideline. Searches were restricted to certain study designs if specified in the
review protocol, and conducted in the following databases:
• Australian Education Index (AEI)

• Applied Social Services Index and Abstracts (ASSIA)
• British Education Index (BEI)
• Cumulative Index to Nursing and Allied Health Literature (CINAHL)

• Cochrane Database of Abstracts of Reviews of Effects (DARE)
• Cochrane Database of Systematic Reviews (CDSR)
• CENTRAL
• Education Resources in Curriculum (ERIC)
• Embase
• HTA database (technology assessments)
• International Bibliography of Social Science (IBSS)
• MEDLINE/MEDLINE In-Process
• Psychological Information Database (PsycINFO)
• Social Services Abstracts (SSA)
• Sociological Abstracts.
The search strategies were initially developed for MEDLINE before being translated
for use in other databases/interfaces. Strategies were built up through a number of
trial searches and discussions of the results of the searches with the review team and
GDG to ensure that all possible relevant search terms were covered. The search
terms for each search are set out in full in Appendix 13.
Reference management
Citations from each search were downloaded into reference management software
and duplicates removed. Records were then screened against the eligibility criteria
of the reviews before being appraised for methodological quality (see below). The
unfiltered search results were saved and retained for future potential re-analysis to
help keep the process both replicable and transparent.
Search filters
To aid retrieval of relevant and sound studies, filters were used to limit a number of
searches to systematic reviews, RCTs and qualitative studies. The search filters for
systematic reviews and RCTs are adaptations of filters designed by the CRD and the
Health Information Research Unit of McMaster University, Ontario. The qualitative
research filter was developed in-house. Each filter comprises index terms relating to
the study type(s) and associated text-words for the methodological description of the
design(s).
Date and language restrictions

Systematic database searches were initially conducted in June 2012 up to the most
recent searchable date. Search updates were generated on a 6-monthly basis, with
the final re-runs carried out in June 2013 to October 2013 ahead of the guideline
consultation. After this point, studies were only included if they were judged by the
GDG to be exceptional (for example, if the evidence was likely to change a
recommendation).
Although no language restrictions were applied at the searching stage, foreign

language papers were not requested or reviewed, unless they were of particular
importance to a review question.

Date restrictions were not applied, except for updates of systematic reviews which
were limited to the date the last searches were conducted. Searches for systematic
reviews and qualitative research were also restricted to a shorter time frame as older
research was thought to be less useful.
Other search methods

Other search methods involved: (a) scanning the reference lists of all eligible
publications (systematic reviews, stakeholder evidence and included studies) for
more published reports and citations of unpublished research; (b) sending lists of
studies meeting the inclusion criteria to subject experts (identified through searches
and the GDG) and asking them to check the lists for completeness, and to provide
information of any published or unpublished research for consideration (see
Appendix 5); (c) checking the tables of contents of key journals for studies that might
have been missed by the database and reference list searches; (d) tracking key papers
in the Science Citation Index (prospectively) over time for further useful references;
(e) conducting searches in ClinicalTrials.gov for unpublished trial reports; (f)
contacting included study authors for unpublished or incomplete datasets. Searches
conducted for existing NICE guidelines were updated where necessary. Other
relevant guidelines were assessed for quality using the AGREE instrument (AGREE
Collaboration, 2003). The evidence base underlying high-quality existing guidelines
was utilised and updated as appropriate.
Full details of the search strategies and filters used for the systematic review of
clinical evidence are provided in Appendix 13.
Study selection and assessment of methodological quality

All primary-level studies included after the first scan of citations were acquired in
full and re-evaluated for eligibility at the time they were being entered into the study
information database. More specific eligibility criteria were developed for each
review question and are described in the relevant clinical evidence chapters. Eligible
systematic reviews and primary-level studies were critically appraised for
methodological quality (risk of bias) using a checklist (see The Guidelines Manual
(NICE, 2012b) for templates). The eligibility of each study was confirmed by at least
one member of the GDG.
For some review questions, it was necessary to prioritise the evidence with respect to
the UK context (that is, external validity). To make this process explicit, the GDG
took into account the following factors when assessing the evidence:
• participant factors (for example, gender, age and ethnicity)

• provider factors (for example, model fidelity, the conditions under which the
intervention was performed and the availability of experienced staff to
undertake the procedure)
• cultural factors (for example, differences in standard care and differences in
the welfare system).

It was the responsibility of the GDG to decide which prioritisation factors were
relevant to each review question in light of the UK context.
Unpublished evidence
Stakeholders, authors and principle investigators were approached for unpublished
evidence (see Appendix 5). The GDG used a number of criteria when deciding
whether or not to accept unpublished data. First, the evidence must have been
accompanied by a trial report containing sufficient detail to properly assess risk of
bias. Second, the evidence must have been submitted with the understanding that
data from the study and a summary of the study’s characteristics would be
published in the full guideline. Therefore, in most circumstances the GDG did not
accept evidence submitted ‘in confidence’. However, the GDG recognised that
unpublished evidence submitted by investigators might later be retracted by those
investigators if the inclusion of such data would jeopardise publication of their
research.
Experience of care
Reviews were sought of qualitative studies that used relevant first-hand experiences
of carers. The experience of service users with mental health problems has been
reviewed in Service User Experience in Adult Mental Health (NCCMH, 2012 [full
guideline]). Therefore, for this guideline, only a review of the carer experience of
care was conducted. A particular outcome was not specified by the GDG. Instead,
the review was concerned with narrative data that highlighted the experience of
care. Where the search did not generate an adequate body of literature, a further
search for primary qualitative studies was undertaken.
3.5.2 Data extraction

Quantitative analysis
Study characteristics, aspects of methodological quality, and outcome data were
extracted from all eligible studies, using Review Manager 5.1 (The Cochrane
Collaboration, 2011) and an Excel-based form (see Appendix 7).
In most circumstances, for a given outcome (continuous and dichotomous), where

more than 50% of the number randomised to any group were missing or incomplete,
the study results were excluded from the analysis (except for the outcome ‘leaving
the study early’, in which case, the denominator was the number randomised).
Where there were limited data for a particular review, the 50% rule was not applied.
In these circumstances the evidence was downgraded (see section 3.5.4).
Where possible, outcome data from an intention-to-treat analysis (ITT) (that is, a
‘once-randomised-always-analyse’ basis) were used. Where ITT had not been used
or there were missing data, the effect size for dichotomous outcomes were
recalculated using best-case and worse-case scenarios. Where conclusions varied
between scenarios, the evidence was downgraded (see section 3.5.4).

Where some of the studies failed to report standard deviations (for a continuous
outcome), and where an estimate of the variance could not be computed from other
reported data or obtained from the study author, the following approach was
taken. 3When the number of studies with missing standard deviations was less than
one-third and when the total number of studies was at least ten, the pooled standard
deviation was imputed (calculated from all the other studies in the same meta-
analysis that used the same version of the outcome measure). In this case, the
appropriateness of the imputation was made by comparing the standardised mean
differences (SMDs) of those trials that had reported standard deviations against the
hypothetical SMDs of the same trials based on the imputed standard deviations. If
they converged, the meta-analytical results were considered to be reliable.
When the conditions above could not be met, standard deviations were taken from
another related systematic review (if available). In this case, the results were
considered to be less reliable.
The meta-analysis of survival data, such as time to any mood episode, was based on
log hazard ratios and standard errors. Since individual participant data were not
available in included studies, hazard ratios and standard errors calculated from a
Cox proportional hazard model were extracted. Where necessary, standard errors
were calculated from confidence intervals (CIs) or p value according to standard
formulae (see the Cochrane Reviewers’ Handbook5.1.0 (Higgins & Green)). Data
were summarised using the generic inverse variance method using Review
Manager.
Consultation with another reviewer or members of the GDG was used to overcome
difficulties with coding. Data from studies included in existing systematic reviews
were extracted independently by one reviewer and cross-checked with the existing
dataset. Where possible, two independent reviewers extracted data from new
studies. Where double data extraction was not possible, data extracted by one
reviewer was checked by the second reviewer. Disagreements were resolved
through discussion. Where consensus could not be reached, a third reviewer or GDG
members resolved the disagreement. Masked assessment (that is, blind to the journal
from which the article comes, the authors, the institution and the magnitude of the
effect) was not used since it is unclear that doing so reduces bias (Berlin, 2001; Jadad
et al., 1996).
Qualitative analysis
After transcripts/reviews or primary studies of carer experience were identified (see
3.5.1), each was read and re-read and sections of the text were collected under
different headings. Under the broad headings, specific emergent themes were
identified and coded by two researchers working independently. Overlapping
themes and themes with the highest frequency count across all testimonies were
3Based on the approach suggested by Furukawa and colleagues (2006).

extracted and regrouped. The findings from this qualitative analysis can be found in
Chapter 4.
The quality of the included studies was assessed using the NICE quality checklist for
qualitative literature (see The Guidelines Manual (NICE, 2012b) for templates). The
domains of this checklist (including the theoretical approach, study design, validity
and data analysis) aim to provide a transparent description of methods in order to
assess the reliability and transferability of the findings of primary studies to their
setting. As there is currently no accepted gold standard of assessing study quality,
studies were not excluded or weighted on the basis of quality.
3.5.3 Evidence synthesis

The method used to synthesize evidence depended on the review question and
availability and type of evidence (see Appendix 6 for full details). Briefly, for
questions about the psychometric properties of instruments, reliability, validity and
clinical utility were synthesized narratively based on accepted criteria. For questions
about test accuracy, bivariate test accuracy meta-analysis was conducted where
appropriate. For questions about the effectiveness of interventions, standard meta-
analysis or network meta-analysis was used where appropriate, otherwise narrative
methods were used with clinical advice from the GDG. In the absence of high-
quality research, an informal consensus process was used (see 3.5.7).
3.5.4 Grading the quality of evidence

For questions about the effectiveness of interventions, the GRADE approach 4 was
used to grade the quality of evidence for each outcome (Guyatt et al., 2011). For
questions about the experience of care and the organisation and delivery of care,
methodology checklists (see section 3.5.1) were used to assess the risk of bias, and
this information was taken into account when interpreting the evidence. The
technical team produced GRADE evidence profiles (see below) using GRADE
profiler (GRADEpro) software (Version 3.6), following advice set out in the GRADE
handbook (Schünemann et al., 2009). Those doing GRADE ratings were trained, and
calibration exercises were used to improve reliability (Mustafa et al., 2013).
Evidence profiles
A GRADE evidence profile was used to summarise both the quality of the evidence
and the results of the evidence synthesis for each ‘critical’ and ‘important’ outcome
(see Table 3 for an example of an evidence profile). The GRADE approach is based
on a sequential assessment of the quality of evidence, followed by judgment about
the balance between desirable and undesirable effects, and subsequent decision
about the strength of a recommendation.
Within the GRADE approach to grading the quality of evidence, the following is
used as a starting point:
4 For further information about GRADE, see www.gradeworkinggroup.org

• RCTs without important limitations provide high quality evidence
• observational studies without special strengths or important limitations
provide low quality evidence.
For each outcome, quality may be reduced depending on five factors:

methodological limitations, inconsistency, indirectness, imprecision and publication
bias. For the purposes of the guideline, each factor was evaluated using criteria
provided in Table 4.
For observational studies without any reasons for down-grading, the quality may be
up-graded if there is a large effect, all plausible confounding would reduce the
demonstrated effect (or increase the effect if no effect was observed), or there is
evidence of a dose-response gradient (details would be provided under the ‘other’
column).
Each evidence profile includes a summary of findings: number of participants

included in each group, an estimate of the magnitude of the effect, and the overall
quality of the evidence for each outcome. Under the GRADE approach, the overall
quality for each outcome is categorised into one of four groups (high, moderate, low,
very low).

Table 3: Example of a GRADE evidence profile
Quality assessment No of patients Effect

Quality Importance
Other
No of Intervent Control Relative
Design Risk of bias Inconsistency Indirectness Imprecision consider- Absolute
studies ion group (95% CI)
ations
Outcome 1 (measured with: any valid method; Better indicated by lower values)
2 randomi no serious no serious no serious serious1 none 47 43 - SMD 0.20 lower ⊕⊕⊕Ο CRITICAL
sed trials risk of bias inconsistency indirectness (0.61 lower to MODERATE
0.21 higher)
Outcome 2 (measured with: any valid rating scale; Better indicated by lower values)
4 randomi serious2 no serious no serious serious1 none 109 112 - SMD 0.42 lower ⊕⊕ΟΟ CRITICAL
sed trials inconsistency indirectness (0.69 to 0.16 LOW
lower)
26 randomi no serious serious3 no serious no serious none 521/5597 798/3339 RR 0.43 136 fewer per ⊕⊕⊕Ο CRITICAL
sed trials risk of bias indirectness imprecision (9.3%) (23.9%) (0.36 to 1000 (from 117 MODERATE
0.51) fewer to 153
fewer)
5 randomi no serious no serious no serious no serious none 503 485 - SMD 0.34 lower ⊕⊕⊕⊕ CRITICAL
sed trials risk of bias inconsistency indirectness imprecision (0.67 to 0.01 HIGH
lower)
1 Optimal information size (for dichotomous outcomes, OIS = 300 events; for continuous outcomes, OIS = 400 participants) not met.
2 Risk of bias across domains was generally high or unclear.
3 There is evidence of moderate heterogeneity of study effect sizes.

Table 4: Factors that decrease quality of evidence
Factor Description Criteria
Limitations Methodological quality/ risk of Serious risks across most studies (that reported
bias. a particular outcome). The evaluation of risk of
bias was made for each study using NICE
methodology checklists (see Section 3.5.1).
Inconsistency Unexplained heterogeneity of Moderate or greater heterogeneity (see
results. (Schünemann et al., 2009) for further
information about how this was evaluated)
Indirectness How closely the outcome If the comparison was indirect, or if the
measures, interventions and question being addressed by the GDG was
participants match those of substantially different from the available
interest. evidence regarding the population,
intervention, comparator, or an outcome.
Imprecision Results are imprecise when If either of the following two situations were
studies include relatively few met:
patients and few events and thus • the optimal information size (for
have wide confidence intervals dichotomous outcomes, OIS = 300
around the estimate of the effect. events; for continuous outcomes, OIS =
400 participants) was not achieved
• the 95% confidence interval around the
pooled or best estimate of effect
included both 1) no effect and 2)
appreciable benefit or appreciable harm
Publication Systematic underestimate or an Evidence of selective publication. This may be
bias overestimate of the underlying detected during the search for evidence, or
beneficial or harmful effect due to through statistical analysis of the available
the selective publication of evidence.
studies.
3.5.5 Presenting evidence to the Guideline Development Group

Study characteristics tables and, where appropriate, forest plots generated with
Review Manager Version 5.2 and GRADE summary of findings tables (see below)
were presented to the GDG.
Where meta-analysis was not appropriate and/or possible, the reported results from
each primary-level study were included in the study characteristics table. The range
of effect estimates were included in the GRADE profile, and where appropriate,
described narratively.
Summary of findings tables

Summary of findings tables generated from GRADEpro were used to summarise the
evidence for each outcome and the quality of that evidence (Table 5). The tables
provide illustrative comparative risks, especially useful when the baseline risk varies
for different groups within the population.
Psychosis & schizophrenia in adults 55

Table 5: Example of a GRADE summary of findings table
Patient or population:
Settings:
Intervention:
Comparison:
Outcomes Illustrative comparative risks* (95% Relative No of Quality of Comments
CI) effect Participants the evidence
Assumed risk Corresponding risk (95% CI) (studies) (GRADE)
Any control Intervention group
group
Outcome 1 The mean outcome in 90 ⊕⊕⊕⊝
any valid the intervention (2 studies) moderate1
rating scale group was
0.20 standard
deviations lower
(0.61 lower to 0.21
higher)
Outcome 2 The mean outcome in 221 ⊕⊕⊝⊝
any valid the intervention (4 studies) low1,2
0.42 standard
deviations lower
(0.69 to 0.16 lower)
Outcome 3 239 per 1000 103 per 1000 RR 0.43 8936 ⊕⊕⊕⊝
any valid (86 to 122) (0.36 to (26 studies) moderate3
rating scale 0.51)
Outcome 4 The mean outcome in 988 ⊕⊕⊕⊕
any valid the intervention (5 studies) high
0.34 standard
deviations lower
(0.67 to 0.01 lower)
*The basis for the assumed risk (for example, the median control group risk across studies) is provided
in the footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk
in the comparison group and the relative effect of the intervention (and its 95% CI).
Note. CI = Confidence interval.
1 Optimal information size (for dichotomous outcomes, OIS = 300 events; for continuous outcomes,
OIS = 400 participants) not met.

2 Risk of bias across domains was generally high or unclear.
3 There is evidence of moderate heterogeneity of study effect sizes.

3.5.6 Extrapolation
When answering review questions, if there is no direct evidence from a primary
dataset 5 based on the initial search for evidence it may be appropriate to extrapolate
from another dataset. In this situation, the following principles were used to
determine when to extrapolate:
• a primary dataset is absent, of low quality or is judged to be not relevant to
the review question under consideration
• a review question is deemed by the GDG to be important, such that in the
absence of direct evidence, other data sources should be considered
• non-primary data source(s) is in the view of the GDG available, which may
inform the review question.
When the decision to extrapolate was made, the following principles were used to
inform the choice of the non-primary dataset:
• the populations (usually in relation to the specified diagnosis or problem
which characterises the population) under consideration share some common
characteristic but differ in other ways, such as age, gender or in the nature of
the disorder (for example, a common behavioural problem; acute versus
chronic presentations of the same disorder); and
• the interventions under consideration in the view of the GDG have one or
more of the following characteristics:
- share a common mode of action (for example, the pharmacodynamics of
drug; a common psychological model of change - operant conditioning)
- be feasible to deliver in both populations (for example, in terms of the
required skills or the demands of the health care system)
- share common side effects/harms in both populations; and
• the context or comparator involved in the evaluation of the different datasets
shares some common elements which support extrapolation; and
• the outcomes involved in the evaluation of the different datasets shares some
common elements which support extrapolation (for example, improved mood
or a reduction in challenging behaviour).
When the choice of the non-primary dataset was made, the following principles
were used to guide the application of extrapolation:
• the GDG should first consider the need for extrapolation through a review of
the relevant primary dataset and be guided in these decisions by the
principles for the use of extrapolation
• in all areas of extrapolation datasets should be assessed against the principles
for determining the choice of datasets. In general the criteria in the four
principles set out above for determining the choice should be met
• in deciding on the use of extrapolation, the GDG will have to determine if the
extrapolation can be held to be reasonable, including ensuring that:
5A primary dataset is defined as a dataset which contains evidence on the population and intervention under
review

- the reasoning behind the decision can be justified by the clinical need for a
recommendation to be made
- the absence of other more direct evidence, and by the relevance of the
potential dataset to the review question can be established
- the reasoning and the method adopted is clearly set out in the relevant
section of the guideline.
3.5.7 Method used to answer a review question in the absence of

appropriately designed, high-quality research
In the absence of appropriately designed, high-quality research (including indirect
evidence where it would be appropriate to use extrapolation), an informal consensus
process was adopted. The process involved a group discussion of what is known
about the issues. The views of GDG were synthesised narratively by a member of the
review team, and circulated after the meeting. Feedback was used to revise the text,
which was then included in the appropriate evidence review chapter.
3.6 HEALTH ECONOMICS METHODS

The aim of the health economics was to contribute to the guideline’s development by
providing evidence on the cost effectiveness of interventions for adults with
psychosis and schizophrenia covered in the guideline. This was achieved by:
• systematic literature review of existing economic evidence

• decision-analytic economic modelling.
Systematic reviews of economic literature were conducted in all areas covered in the
guideline. Economic modelling was undertaken in areas with likely major resource
implications, where the current extent of uncertainty over cost effectiveness was
significant and economic analysis was expected to reduce this uncertainty, in
accordance with The Guidelines Manual(NICE, 2012b). Prioritisation of areas for
economic modelling was a joint decision between the Health Economist and the
GDG. The rationale for prioritising review questions for economic modelling was set
out in an economic plan agreed between NICE, the GDG, the Health Economist and
the other members of the technical team. For the 2014 guideline, the cost
effectiveness of vocational rehabilitation for people with psychosis and
schizophrenia was selected as a key issue that was addressed by economic
modelling.
In addition, literature on the health-related quality of life of people with psychosis

and schizophrenia was systematically searched to identify studies reporting
appropriate utility scores that could be utilised in a cost-utility analysis.
The rest of this section describes the methods adopted in the systematic literature
review of economic studies. Methods employed in economic modelling are
described in the respective sections of the guideline.

3.6.1 Search strategy for economic evidence
Scoping searches
A broad preliminary search of the literature was undertaken in August 2011to obtain
an overview of the issues likely to be covered by the scope, and help define key
areas. Searches were restricted to economic studies and HTA reports, and conducted
in the following databases:
• Embase
• NHS Economic Evaluation Database (NHS EED)
Any relevant economic evidence arising from the clinical scoping searches was also
made available to the health economist during the same period.
Systematic literature searches

After the scope was finalised, a systematic search strategy was developed to locate
all the relevant evidence. Searches were restricted to economic studies and health
technology assessment reports, and conducted in the following databases:
• Embase
• NHS EED
• PsycINFO
Any relevant economic evidence arising from the clinical searches was also made
available to the health economist during the same period.
The search strategies were initially developed for MEDLINE before being translated
for use in other databases/interfaces. Strategies were built up through a number of
trial searches, and discussions of the results of the searches with the review team and
GDG to ensure that all possible relevant search terms were covered. In order to
assure comprehensive coverage, search terms for the population were kept
purposely broad to help counter dissimilarities in database indexing practices and
thesaurus terms, and imprecise reporting of study populations by authors in the
titles and abstracts of records.
For standard mainstream bibliographic databases (Embase, MEDLINE and

PsycINFO) search terms were combined with a search filter for health economic
studies. For searches generated in topic-specific databases (HTA, NHS EED) search
terms were used without a filter. The search terms are set out in full in Appendix 14.

Reference management
Citations from each search were downloaded into reference management software
and duplicates removed. Records were then screened against the inclusion criteria of
the reviews before being quality appraised. The unfiltered search results were saved
and retained for future potential re-analysis to help keep the process both replicable
and transparent.
Search filters
The search filter for health economics is an adaptation of a pre-tested strategy
designed by CRD (2007). The search filter is designed to retrieve records of economic
evidence (including full and partial economic evaluations) from the vast amount of
literature indexed to major medical databases such as MEDLINE. The filter, which
comprises a combination of controlled vocabulary and free-text retrieval methods,
maximises sensitivity (or recall) to ensure that as many potentially relevant records
as possible are retrieved from a search. A full description of the filter is provided in
Appendix 14.
Date and language restrictions

Systematic database searches were initially conducted in June 2012up to the most
recent searchable date. Search updates were generated on a 6-monthly basis, with
the final re-runs carried out in June 2013 ahead of the guideline consultation. After
this point, studies were included only if they were judged by the GDG to be
exceptional (for example, the evidence was likely to change a recommendation).
Although no language restrictions were applied at the searching stage, foreign

language papers were not requested or reviewed, unless they were of particular
importance to an area under review. In order to obtain data relevant to current
healthcare settings and costs, all the searches were restricted to research published
from 1996 onwards, except for an update search of an existing review from Chapter
5, which was limited from the date the last search was conducted.
Other search methods

Other search methods involved scanning the reference lists of all eligible
publications (systematic reviews, stakeholder evidence and included studies from
the economic and clinical reviews) to identify further studies for consideration.
Full details of the search strategies and filter used for the systematic review of health
economic evidence are provided in Appendix 14.
3.6.2 Inclusion criteria for economic studies

The following inclusion criteria were applied to select studies identified by the
economic searches for further consideration:
1. Only English language papers were considered.

2. Only studies from Organisation for Economic Co-operation and Development
countries were included, as the aim of the review was to identify economic
information transferable to the UK context.
3. Studies published from 2002 onwards were included. This date restriction
was imposed to obtain data relevant to current healthcare settings and costs.
4. Selection criteria based on types of clinical conditions and service users as
well as interventions assessed were identical to the clinical literature review.
5. Studies were included provided that sufficient details regarding methods and
results were available to enable the methodological quality of the study to be
assessed, and provided that the study’s data and results were extractable.
Poster presentations, abstracts, dissertations, commentaries and discussion
publications were excluded.
6. Full economic evaluations that compared two or more relevant interventions
and considered both costs and consequences, as well as costing analyses
comparing only costs between two or more interventions, were included in
the review.
7. Economic studies were included if they used clinical effectiveness data from
an RCT, a prospective cohort study, pre- and post-observational studies or a
systematic review and meta-analysis of clinical studies. Studies that utilised
clinical effectiveness parameters based mainly on expert opinion or
assumptions were excluded from the review.
8. Studies were included only if the examined interventions and populations
under consideration were clearly described.
9. Studies that adopted a very narrow perspective, ignoring major categories of
costs relevant to the NHS, were excluded; for example studies that estimated
exclusively hospitalisation costs were considered non-informative to the
guideline development process. Also, studies that considered other types of
costs, except direct healthcare costs, were excluded from this review.
3.6.3 Applicability and quality criteria for economic studies

All economic papers eligible for inclusion were appraised for their applicability and
quality using the methodology checklist for economic evaluations recommended by
NICE (NICE, 2012b). The methodology checklist for economic evaluations was also
applied to the economic models developed specifically for this guideline. All studies
that fully or partially met the applicability and quality criteria described in the
methodology checklist were considered during the guideline development process,
along with the results of the economic modelling conducted specifically for this
guideline. The completed methodology checklists for all economic evaluations
considered in the guideline are provided in Appendix 18.
3.6.4 Presentation of economic evidence

The economic evidence considered in the guideline is provided in the respective
evidence chapters, following presentation of the relevant clinical evidence. The
references to included studies and the respective evidence tables with the study
characteristics and results are provided in Appendix 19. Methods and results of

economic modelling undertaken alongside the guideline development process are
presented in the relevant evidence chapters. Characteristics and results of all
economic studies considered during the guideline development process (including
modelling studies conducted for this guideline) are summarised in economic
evidence profiles accompanying respective GRADE clinical evidence profiles in
Appendix 17.
3.6.5 Results of the systematic search of economic literature

The titles of all studies identified by the systematic search of the literature were
screened for their relevance to the topic (that is, economic issues and information on
health-related quality of life in people with psychosis and schizophrenia). References
that were clearly not relevant were excluded first. The abstracts of all potentially
relevant studies (90 references) were then assessed against the inclusion criteria for
economic evaluations by the health economist. Full texts of the studies potentially
meeting the inclusion criteria (including those for which eligibility was not clear
from the abstract) were obtained. Studies that did not meet the inclusion criteria,
were duplicates, were secondary publications of one study, or had been updated in
more recent publications were subsequently excluded. Economic evaluations eligible
for inclusion (47 references) were then appraised for their applicability and quality
using the methodology checklist for economic evaluations. Finally, 21 economic
studies identified by the systematic literature search, as well as two studies that were
unpublished at the time of the guideline development and were identified through
consultation with the GDG, met fully or partially the applicability and quality
criteria for economic studies, and were thus considered at formulation of the
guideline recommendations.
3.7 LINKING EVIDENCE TO RECOMMENDATIONS

Once the clinical and health economic evidence was summarised, the GDG drafted
the recommendations. In making recommendations, the GDG took into account the
trade-off between the benefits and harms of the intervention/instrument, as well as
other important factors, such as economic considerations, values of the GDG and
society, the requirements to prevent discrimination and to promote equality 6, and
the GDG’s awareness of practical issues (Eccles et al., 1998; NICE, 2012b).
Finally, to show clearly how the GDG moved from the evidence to the
recommendations, each chapter has a section called ‘linking evidence to
recommendations’. Underpinning this section is the concept of the ‘strength’ of a
recommendation (Schünemann et al., 2003). This takes into account the quality of the
evidence but is conceptually different. Some recommendations are ‘strong’ in that
the GDG believes that the vast majority of healthcare professionals and service users
would choose a particular intervention if they considered the evidence in the same
way that the GDG has. This is generally the case if the benefits clearly outweigh the
harms for most people and the intervention is likely to be cost effective. However,
6See NICE’s equality scheme: www.nice.org.uk/aboutnice/howwework/NICEEqualityScheme.jsp

there is often a closer balance between benefits and harms, and some service users
would not choose an intervention whereas others would. This may happen, for
example, if some service users are particularly averse to some side effect and others
are not. In these circumstances the recommendation is generally weaker, although it
may be possible to make stronger recommendations about specific groups of service
users. The strength of each recommendation is reflected in the wording of the
recommendation, rather than by using ratings, labels or symbols.
Where the GDG identified areas in which there are uncertainties or where robust
evidence was lacking, they developed research recommendations. Those that were
identified as ‘high priority’ were developed further in the NICE version of the
guideline, and presented in Appendix 10.
3.8 STAKEHOLDER CONTRIBUTIONS

Professionals, service users, and companies have contributed to and commented on
the guideline at key stages in its development. Stakeholders for this guideline
include:
• service user and carer stakeholders: national service user and carer
organisations that represent the interests of people whose care will be covered
by the guideline
• local service user and carer organisations: but only if there is no relevant
national organisation
• professional stakeholders’ national organisations: that represent the
healthcare professionals who provide the services described in the guideline
• commercial stakeholders: companies that manufacture drugs or devices used
in treatment of the condition covered by the guideline and whose interests
may be significantly affected by the guideline
• providers and commissioners of health services in England and Wales
• statutory organisations: including the Department of Health, the Welsh
Assembly
• Government, NHS Quality Improvement Scotland, the Care Quality
Commission and the National Patient Safety Agency
• research organisations: that have carried out nationally recognised research in
the area.
NICE clinical guidelines are produced for the NHS in England and Wales, so a
‘national’ organisation is defined as one that represents England and/or Wales, or
has a commercial interest in England and/or Wales.
Stakeholders have been involved in the guideline’s development at the following

points:
• commenting on the initial scope of the guideline and attending a scoping

workshop held by NICE
• contributing possible review questions and lists of evidence to the GDG
• commenting on the draft of the guideline.

3.9 VALIDATION OF THE GUIDELINE
Registered stakeholders had an opportunity to comment on the draft guideline,
which was posted on the NICE website during the consultation period. Following
the consultation, all comments from stakeholders and experts (see Appendix 4B)
were responded to, and the guideline updated as appropriate. NICE also reviewed
the guideline and checked that stakeholders' comments had been addressed.
Following the consultation period, the GDG finalised the recommendations and the
NCCMH produced the final documents. These were then submitted to NICE for a
quality assurance check. Any errors were corrected by the NCCMH, then the
guideline was formally approved by NICE and issued as guidance to the NHS in
England and Wales.

4 CARERS’ EXPERIENCE
4.1 INTRODUCTION
This chapter is new for the 2014 guideline and aims to evaluate and discuss the
experience of health and social care services of carers of people with severe mental
illness, including psychosis and schizophrenia (see Section 4.2). The chapter also
evaluates the effectiveness of interventions that aim to improve carers’ experience of
caring and of services (see Section 4.3). The GDG has sought to identify and evaluate
factors and attributes of health and social care services that positively or negatively
affect the carers’ experiences of services and what can be done by health and social
care services to improve the experience of services and the wellbeing of carers. For
the purposes of this guideline, ‘carers’ are defined as family and friends who may or
may not live with the service user, and who provide informal and regular care and
support to someone with a severe mental illness such as psychosis and
schizophrenia.
The population of interest in this chapter is carers of people with severe mental
illness, including psychosis and schizophrenia. Service user experience of the
treatment and management of these conditions in adult mental health services has
been comprehensively reviewed in Service User Experience in Adult Mental Health
(NICE, 2011). Therefore it is important that this chapter is taken in conjunction with
that guidance because service user experience is not the focus of this review.
In the UK just over half of people with schizophrenia are in contact with a close
relative of whom 65% will be female and 36% a parent (Roick et al., 2007). It is
important to acknowledge that caring can be a strongly positive experience.
Nevertheless, most who write about it describe the impact in terms of a ‘burden’ that
is both subjective (perceived) and objective (for example, contributing directly to ill
health and financial problems or in displacing other daily routines) (Awad &
Voruganti, 2008), and varies between different cultures (Rosenfarb et al., 2006). A
European study (based in Italy, England, Germany, Greece and Portugal) reported
that carers for adults with schizophrenia spent an average of 6 to 9 hours per day
providing care (Magliano et al., 1998). Many people are not able to work or have to
take time off work to provide care, and when these costs are combined with those of
replacing carers with paid workers, the annual estimate of the potential cost to the
NHS is £34,000 per person with schizophrenia (Andrew et al., 2012).
Supporting carers can be very challenging and it is sometimes difficult for health
and social care professionals to identify what carers find the most helpful at different
stages of the care pathway. Information and support that is offered at the early
stages of care can be the most effective, particularly if it provides a sound base of
knowledge and skills from which carers can draw upon at different times. It is
recognised that families and friends can either help or a hinder the recovery of
service user, but some interventions, such as family intervention, have a substantial

impact on relapse rates (see Chapter 9 which gives an account of this and shows the
beneficial effects of family intervention for the families of people with psychosis and
schizophrenia). However, these interventions remain difficult to access (Fadden &
Heelis, 2011). At times of crisis the needs of carers are much more urgent; therefore
easy access to supportive allies can be very helpful at these times.
European studies of the relatives of people with schizophrenia showed that the
burden of care was lower when psychosocial interventions were provided to service
users and their relatives and professional and social network support was available
(Jeppesen et al., 2005; Magliano et al., 2006). Information sharing and the issue of
confidentiality is a particular concern of people with psychosis and schizophrenia
and their families and carers because of the sensitive nature of mental health
problems, which is compounded by differences of opinion held by professionals
about what information can be shared. This contrasts with clinical practice in other
areas of health where increasingly the emphasis is on healthcare being seen as a
partnership between professionals, service users and their families and carers, based
on appropriate sharing of information. In its guidance Carers and Confidentiality, the
Royal College of Psychiatrists has recognised the importance of training
practitioners in confidentiality and information sharing to empower service users
and their carers (Royal College of Psychiatrists, 2010).
Current practice
There are huge variations in the provision of family intervention or other support for
carers and in the extent to which professionals appreciate the important role of
carers in the lives and recovery of many (but not all) service users. Moreover,
professionals are often confused about issues such as confidentiality and information
sharing, leaving carers often feeling isolated and alone. Many carers therefore turn to
voluntary sector organisations such as ‘Rethink’. As a result there is not a consistent
approach to health and social care support to carers across the country. In some
areas carers are well supported through mental health services, although this is
probably the exception. Carers are often unsure about their role or even about their
rights, such as the right to a carers’ assessment. The 2002 and 2009 guidelines if not
fully address these needs and evaluate more precisely the needs of carers.
This chapter attempts to redress this imbalance in two ways. First, the GDG has
conducted a review of qualitative studies of carers’ experiences of health and social
care services. Second, the GDG decided to search for and evaluate quantitative trials
of interventions specifically aimed at improving the experience of carers.
4.2 CARERS’ EXPERIENCE (QUALITATIVE REVIEW)

4.2.1 Introduction
Definition and aim of review
The aim of this qualitative review is to evaluate the experience of care from the
perspective of informal carers of people with severe mental illness. Specifically, the

review includes studies that focus on factors relating to health and social services
that have a beneficial or detrimental effect on the carers’ overall experience of care.
This qualitative review precedes a review of interventions that examines which

modifications to health and social services improve the experience of using services
for carers of adults with severe mental illness (Section 4.3).
4.2.2 Review protocol (carers’ experience qualitative review)

The review protocol summary, including the review question(s), information about
the databases searched, and the eligibility criteria used for this section of the
guideline, can be found in Table 6 (a complete list of review questions and the full
review protocol can be found in Appendix 6; further information about the search
strategy can be found in Appendix 13).
Table 6: Clinical review protocol summary for the qualitative review of carers’
experience
Component Description
Review question What factors improve or diminish the experience of health and social services
for carers of people with severe mental illness?
Objectives To identify factorsthat improve or diminish carers’ experiences of health and
social servicesand carers’ wellbeing.
Population Included
Carers of adults (18+) and people in early intervention services (which may
include people 14 years and older) with severe mental illness who use health
and social services in community settings.
Include papers with a service user population of at least:

66% schizophrenia or
66% schizophrenia and bipolar disorder or
66% schizophrenia and ‘mood disorders’ or
66% undefined severe mental illness
66% bipolar disorder.
Excluded
Studies conducted in low and middle income countries were excluded as the
service provision is not comparable to the UK.
Intervention(s) Actions by health and social services that could improve or diminish carers’
experience of health and social services for example:
• form, frequency, and content of interactions with carers
• organisation of services and interactions with carers
• sharing information with carers and receiving information from carers.
Comparison N/A
Critical outcomes Themes and specific issues that carers identify as improving or diminishing
their experience of health and social care
Study design • Metasynthesis of qualitative studies including people who care for
people with severe mental illness
• Qualitative primary studies (focus group, semi-structured interviews
and written responses to open-ended questions) including people who
care for people with severe mental illness
NB: Studies that examined the views of carers in addition to other stakeholders
(including healthcare professionals and service users) were only included if the

views of carers were separable from non-carers.
Electronic Core databases:
databases CENTRAL, CDSR, DARE, HTA, Embase, MEDLINE, MEDLINE In-Process
Topic specific databases: AEI, ASSIA, BEI, CINAHL, ERIC, IBSS, PsycINFO,
Sociological Abstracts, SSA
Date searched 2002 to June 2013
The GDG decided that knowledge, understanding and experience of health and
social care prior to 2002 would not be relevant to present day services.
Review strategy Thematic synthesis of qualitative studies.
4.2.3 Method
A systematic review and a narrative thematic synthesis of qualitative studies was
carried out using the methods described by Thomas and Harden (2008) (see Chapter
3 for further information). Quality checklists were completed for all included studies
(see Section 4.2.5 for a summary and Appendix 15b for the full checklists).
4.2.4 Studies considered7

Twenty-six primary studies (N = 695) providing relevant data met the eligibility
criteria for this review: ASKEY2009 (Askey et al., 2009), BARNABLE2006 (Barnable
et al., 2006), BERGNER2008 (Bergner et al., 2008), CHIU2006 (Chiu et al., 2006),
GOODWIN2006 (Goodwin & Happell, 2006), HUGHES2011 (Hughes et al., 2011),
JANKOVIC2011 (Jankovic et al., 2011), KNUDSON2002 (Knudson & Coyle, 2002),
LAIRD2010 (Laird et al., 2010), LEVINE2002 (Levine & Ligenza, 2002), LOBBAN2011
(Lobban et al., 2011), LUMSDEN2011 (Lumsden & Rajan, 2011), MCAULIFFE2009
(McAuliffe et al., 2009), MCCANN2011 (McCann et al., 2011), MCCANN2012
(McCann et al., 2012a), NICHOLLS2009 (Nicholls & Pernice, 2009), NORDBY2010
(Nordby et al., 2010), REID2005 (Reid et al., 2005), RILEY2011 (Riley et al., 2011),
ROONEY2006 (Rooney et al., 2006), SAUNDERS2002 (Saunders & Byrne, 2002),
SMALL2010 (Small et al., 2010), TANSKANEN2011 (Tanskanen et al., 2011),
TRANVAG2008 (Tranvag & Kristoffersen, 2008), WAINWRIGHT (Wainwright et al.,
In press), WEIMAND2011 (Weimand et al., 2011). Of the included studies, all but
one were published in peer-reviewed journals between 2002 and 2011. Further
information about excluded studies can be found in Appendix 15a.
Of the 26 included studies, 10 were conducted in the UK. The remaining studies
were conducted in Australia (k = 6), Norway (k = 3), the USA (k = 3), New Zealand
(k = 2), Canada (k = 1) and Hong Kong and Taiwan (k = 1). Table 7 provides an
overview of the included studies.
7Here and elsewhere in the guideline, each study considered for review is referred to by a study ID in capital
letters (primary author and date of study publication, except where a study is in press or only submitted for
publication, then a date is not used).

Table 7: Study characteristics table for qualitative studies of carers’ experience
Study ID and Country N Relationship % living Service user Mean % % Principal Data Analysis
year to service with diagnosis age female white experience collection
user service (years) explored
user
ASKEY2009 UK 22 NR 45% Psychosis 51 72% 59% Needs from Focus groups Thematic analysis
mental health and semi-
services structured
interviews
BARNABLE2006 Canada 6 Siblings NR Schizophrenia NR NR NR Life experience Semi- Hermeneutic
with service user structured phenomenology
interviews
BERGNER2008 USA 12 7 mothers NR Schizophrenia 47.8 75% 0% Duration of Individual Thematic analysis
2 fathers spectrum untreated semi-
1 sister disorder psychosis before structured
1 grandmother treatment in interviews
1 uncle service users
with first-episode
psychosis
CHIU2006 Hong 11 4 sisters NR Severe mental NR 90% NR Experiences of Semi- Thematic analysis
Kong and 4 mothers illness the carer structured
Taiwan 2 daughters interviews
1 father
GOODWIN2006 Australia 19 NR NR Consumers of NR NR NR Barriers to Focus groups Content analysis

mental health participation in
services healthcare
HUGHES2011 UK 10 9 parents 40% Schizophrenia 57 90% 80% Experience of Semi- Interpretive
1 sibling assertive structured phenomenological
outreach interviews analysis
JANKOVIC2011 UK 31 16 parents NR 8 schizophrenia NR 61% 67% Experience of Semi- Thematic analysis
7 partners 6 bipolar involuntary structured
4 siblings 7 other psychiatric interviews
2 children psychotic hospital
1 grandmother disorder admission of
1 elderly 1 manic episode their relatives
relative 1 borderline
personality
disorder
1 no mental

illness
2 unavailable
KNUDSON2002 UK 8 6 mothers 62% Schizophrenia 61 75% NR Experience of Semi Thematic analysis
2 fathers caring for a son structured
or daughter with interviews
schizophrenia
LAIRD2010 New 58 Family NR 70% NR NR NR Understanding Semi- Unclear
Zealand members schizophrenia, and opinions on structured
bipolar disorder, the utility of interviews
depression diagnostic labels
LEVINE2002 USA 55 Parents (74%), NR Schizophrenia, 63 NR 100% Identify needs of Focus groups Unclear
spouses, schizoaffective carers (family
siblings and disorder, mood members) of
children disorder or people with
mixture serious mental
illness during a
crisis
LOBBAN2011 UK 23 22 parents NR Psychosis, NR NR 74% Views on design Focus groups Thematic analysis
1 husband bipolar of an educated
tendencies and coping
toolkit for
relative of people
with psychosis
LUMSDEN2011 UK 20 NR NR NR NR 75% 40 % Carer satisfaction Open-ended Unclear
with assertive questionnaire
outreach s self-
completed or
interview
administered
MCAULIFFE2009 Australia 31 16 mothers 25% 96% NR 61% NR Experience and Focus groups Thematic analysis
9 fathers schizophrenia support needs of
3 partners 4.2% bipolar carers of people
3 siblings with severe
mental illness
MCCANN2011 Australia 20 17 parents 90% First episode 49 85% NR Experience of Semi- Interpretive
1 partner psychosis accessing first- structured phenomenological
1 grandparent episode interviews analysis
1 aunt psychosis
services
MCCANN2012 Australia 20 17 parents 90% First episode 49 85% NR Satisfaction with Semi- Interpretive
1 partner psychosis clinicians structured phenomenological

1 grandparent response to them interviews analysis
1 aunt as informal carers
NICHOLLS2009 New 7 6 parents NR 5 schizophrenia NR 100% NR Perceptions of Individual Thematic analysis
Zealand 1 sibling 1 bipolar relationships semi-
1 major with mental structured
depression health interviews
professionals
NORDBY2010 Norway 18 Relatives NR Severe mental NR NR NR Factors that Focus groups Qualitative content
illness contribute to analysis
carers’
participation in
treatment and
rehabilitation of
family members
with severe
mental illness
REID2005 Australia 8 Parents NR Schizophrenia, NR 87% NR Educational Semi- Unclear
bipolar disorder needs of parents structured
or in-depth
schizoaffective interviews
disorder
RILEY2011 UK 12 NR NR First episode NR NR NR Evaluation of an Focus groups Thematic analysis
psychosis educated
programme for
carers
ROONEY2006 Australia 9 NR NR Bipolar NR NR 33% Experience of Semi- Unclear
disorder, carers from structured
schizophrenia, culturally and interviews
major linguistically
depression diverse
backgrounds
SAUNDERS2002 USA 26 NR NR Schizophrenia 59 NR NR Family Postal Thematic analysis
functioning questionnaire
consisting of
open ended
questions
SMALL2010 UK 13 NR NR Schizophrenia NR 54% NR Carers’ burden 3-month Unclear
diaries
combined
with

unstructured
audio- taped
interviews
TANSKANEN201 UK 9 6 mothers NR First episode NR 89% 77% Experiences of Structured Thematic analysis
1 1 sisters psychosis seeking help for interviews
1 partner first episode
1 mother in law psychosis
TRANVAG2008 Norway 8 6 spouses 100% Bipolar affective NR 50% NR Experiences of Individual Ricoeur’s
2 cohabitants disorder living with a semi- phenomenological
partner with structured hermeneutics
bipolar affective interviews
disorder over
time.
WAINWRIGHT UK 23 12 mothers NR Severe mental 59.5 52% 74% Supporting a Focus groups Thematic analysis
10 fathers illness relative in early
1 husband psychosis
WEIMAND2011 Norway 216 156 parents NR NR NR 75% NR Encounters with Questionnaire Content analysis
18 partners mental health (open-ended
27 siblings services questions)
10 children
2 grand-parents
1 foster parent
2 in-laws
Note. NR = Not reported

4.2.5 Quality assessment summary
Table 8 presents specific questions from the quality checklists that are relevant to the
methodology of the studies. Full quality checklists can be found in Appendix 15b.
The methodological quality and potential risk of bias was unclear across studies,
with 12 out of 26 providing insufficient information about the methods employed.
Of these, two (KNUDSON2002, SMALL2010) failed to describe the study objectives
clearly. Seven (GOODWIN2006, KNUDSON2002, LAIRD2010, LUMSDEN2011,
SAUNDERS2002, SMALL2010, WEIMAND2011) provided insufficient information
regarding the rationale for the methodology as well as a justification for sampling
and data analysis methods selected. Details regarding data collection, including a
clear description of the procedure, were insufficiently described in seven studies
(HUGHES2011, KNUDSON2002, LAIRD2010, LUMSDEN2011, SAUNDERS2002,
SMALL2010, WEIMAND2011). Furthermore, 10 studies (ASKEY2009,
GOODWIN2006, HUGHES2011, KNUDSON2002, LAIRD2010, LUMSDEN2011,
SAUNDERS2002, SMALL2010, TRANVAG2008, WEIMAND2011) failed to
adequately describe the reliability of the methodology and/or analysis, such as how
many researchers were involved with data analysis or whether and how any
differences and discrepant results were addressed. Two studies did not provide an
adequate conclusion (LAIRD2010, LEVINE2002) and two (LUMSDEN2011,
SMALL2010) provided only very limited definition of the implications of the study
as well as an adequate consideration of the limitations.
Table 8: Summary of quality assessment

Conclusions adequate
Analysis reliable?
Methods reliable
Clear objectives
Data collection
Study ID
Defensible
ASKEY2009 + + + + ? +
BARNABLE2006 + + + + + +
BERGNER2008 + + + + + +
CHIU2006 + + + + + +
GOODWIN2006 + ? + ? ? +
HUGHES2011 + + ? + + +
JANKOVIC2011 + + + + + +
KNUDSON2002 ? ? ? ? ? +
LAIRD2010 + ? ? ? ? -
LEVINE2002 + + + + + -
LOBBAN2011 + + + + + +
LUMSDEN2011 + ? ? ? ? ?
MCAULIFFE2009 + + + + + +
MCCANN2011 + + + + + +

MCCANN2012 + + + + + +
NICHOLLS2009 + + + + ? +
NORDBY2010 + + + + + +
REID2005 + + + + + +
RILEY2011 + + ? + + +
ROONEY2006 + + + + + +
SAUNDERS2002 + ? ? ? + +
SMALL2010 - ? ? ? ? ?
TANSKANEN2011 + + + + + +
TRANVAG2008 + + ? ? ? +
WAINWRIGHT + + + + + +
WEIMAND2011 + ? ? ? + +
Key: Assessment of these aspects was:
+: Clear/appropriate; -: Unclear/ inappropriate, ?: unsure
4.2.6 Evidence from qualitative studies of carers’ experience of health

and social care services
The findings from this review focus on features of mental health and social care
services that carers believe either improve or diminish their experience of caring for
adults with severe mental illness, including psychosis and schizophrenia. The
review identified five themes: (1) relationships with healthcare providers; (2) valuing
the identity and experience of the carer; (3) sharing decision making and
involvement; (4) providing clear and comprehensible information; and (5) access to
health services. A summary of the findings is presented below.
Relationships with healthcare providers

Carers reported that healthcare professionals who were welcoming, empathic and
interested in the individual needs of carers resulted in a culture of trust, reassurance
and mutual respect. This in turn enabled carers to feel connected with mental health
services and develop an ongoing relationship, which was central to their experience
of care. Building trust and continuous dialogue with healthcare providers was
important for both ensuring and facilitating care for the service users, as well as to
ensure that their own needs as carers were recognised and met. For example, a
sustained connection with healthcare professionals allowed carers to feel that
someone understood their difficulties, which helped to reduce feelings of isolation.
Factors that further enabled this process included healthcare professionals
demonstrating that they were reliable and respectful and also proactively reaching
out to carers to offer support:
Yeah cos if the professional want to contact you, you know they’re going to, whereas if
you have to contact them you might think oh I’m being a nuisance or whatever [group
agreement] so really it needs to come from them…it does, the contact yeah.
(WAINWRIGHT)

Carers often stated that better relationships with healthcare professionals were built
through ease of access to staff who were flexible to the individual needs of the carers
and families:
Simply being there and offering the opportunities. I know I’m 100% confident that I
can pick up the phone and ring any of…[daughter’s name] treating team and I have
done it. I have every confidence in the world that they are there for me.
(MCCANN2011)
In contrast some carers experienced difficulty in accessing healthcare providers and

reflected on their frustration when services failed to provide information or return
telephone calls:
It took a while because no one responded. No one was there, and I had to leave a
message…I was told they would call me, and no one ever called back, or they weren’t
in, so that was the main thing. [They should] just call you back. Ya know, if I’m
calling, ya know, telling you something is going on with my brother, just call back.
(BERGNER2008)
Cooperation between healthcare professionals and carers was also facilitated when
staff listened to the needs and requests of carers and responded appropriately:
I don’t think there is any time that I have voiced my opinion about something that
they haven’t done something about. They always do something about it.
(HUGHES2011)
I was pleasantly surprised by the positive conversation as well as the way we were
received and listened to here. (NORDBY2010)
Conversely carers felt angered and frustrated when healthcare professionals

appeared not to listen to their views and opinions:
Sometimes the professionals don’t listen and understand what’s actually happening
with X. They should listen to what carers are saying more. It makes me feel
frustrated. (ASKEY2009)
Carers also described how a lack of empathy from healthcare professionals

diminished their experience of services. In particular a dismissive attitude from staff
made carers feel undervalued and problematic. These frustrations resulted in
feelings of distrust and undermined collaborative relationships:
I felt that I as a mother was totally ignored from the start. I had to fight and get angry
to be heard. I felt, quite simply, that I was troublesome. (NORDBY2010)
Finally, carers reflected on the difficulty in developing ongoing relationships with

services when they frequently saw different members of the team. Having a single

point of contact and continuity in healthcare providers was therefore highly valued
by some carers.
Valuing the identity and experience of the carer

Prior to contact with services, carers described how they carried the main
responsibility of care for their family member, often in isolation and without external
help. Across the studies contributing to this theme, carers stated how it was
important for healthcare professionals to recognise and acknowledge the roles they
had played in managing the service users’ symptoms and to utilise their acquired
knowledge in the service users’ care plans, for example:
They [carers] suggested that as they knew their relatives well and demonstrated
expertise in their care delivery they should be seen as part of the multidisciplinary
team and respected by professionals. (ASKEY2009)
However, carers described feeling disempowered and alienated when their

expectations of being valued by healthcare professionals were not met. Professionals
were perceived as ignoring and discounting the views of carers and ultimately
appeared arrogant and overconfident:
He [the psychiatrist] wasn’t remotely interested in anything I had to say about my

daughter- he made out that he knew her better than I did. (NICHOLLS2009)
…the shock from putting him in the hospital became so much greater when we
discovered how the system worked. We came with confidence to the professionals; that
they would take care of our son…and that our experiences and knowledge about him
might be useful in the treatment. Instead we experienced to be harshly rejected, in an
almost arrogant manner. (WEIMAND2011)
Carers also felt undervalued and angered when healthcare providers did not
recognise their expertise and apply it to the care of the service user:
You know what is normal for this person. You know what is abnormal. You are the
people who know that and what you say should be taken seriously. This should be
included as part of the initial assessment. (MCAULIFFE2009)
In contrast, carers also identified positive examples with services in which they were
seen as a useful resource and invited to partake in discussions about the service
user’s treatment and care. In these situations, carers described having ‘faith’ in the
system and healthcare professionals, which in turn was associated with a reduced
sense of stress and burden:
At the first time of hospitalization we felt we were excluded and they (i.e. the staff)
had to use their own experiences and would not listen to ours. But this time we have
been invited to tell them about our experiences of his functioning in everyday life at
home. (NORDBY2010)

For carers, the sense of being valued was not solely through having an input into the
service users’ care plan. Healthcare providers acknowledging the carer’s important
role and keeping them informed, where appropriate, also enabled carers to feel
valued.
...the best thing I think was being informed…even if they say, we can’t divulge
anything, it’s still contact, it’s still saying well you are the mum. (REID2005)
Sharing decision making and involvement

The carers’ ability and desire to be actively involved in the service users’ care varied
across studies. However, it was evident that when carers felt informed and
understood the care plan, feelings of anxiety and stress were reduced.
Feeling excluded and increased stress were particularly evident when carers were
unaware of changes to the service users’ treatment plan, which often had
implications for increased responsibility for carers. Lack of information and
opportunities for involvement was largely influenced by the need to balance the
service user’s confidentiality with the carer’s need to be informed. Often carers noted
that members of staff would cite concerns over confidentiality as an explanation for
excluding them from discussions relating to the service user’s care:
We ourselves, really, have been largely side-lined. Uh, things were said ‘Well, these
are now confidential matters’ and, um, we still find that very difficult because, uh,
how can you not be informed about somebody that you’re caring for? Um you need to
know certain things- Otherwise you can’t care properly for that person.
(KNUDSON2002)
Poor communication and lack of involvement led carers to report feeling taken for
granted and unprepared for changes in responsibility. Carers reflected how
healthcare professionals sometimes assumed the carer would automatically take
responsibility without consulting them, which resulted in feelings of anger and
frustration:
One carer related a story about how she was disengaged from discharge planning
discussions only to find that her son was to be discharged to her at a time when she
had arranged to be out of the city visiting a friend. This situation caused a great deal
of trauma for all concerned, and could have been avoided had communication been
more open. (MCAULIFFE2009)
These feelings were heightened when there was disagreement between the carer and
healthcare providers regarding treatment or discharge of the service user:
...we were shattered…I didn’t really want him to come home and spend the night at
home already, and one day I went in and it took me completely by surprise Dr X
wanted him released that day, and I think that [name of service user] had only just
had his first weekend at home…he [name of service user] was being really bolshy and
still very argumentative, and I said you know perhaps we could just sit quietly and

have some time and he was being really horrible…and I really knew I wasn’t ready to
have him home, but it was really obvious that the doctor wanted him to come home
and thought that he was well, and he came home. (JANKOVIC2011)
Carers also provided examples of experiences that fostered effective communication

with healthcare professionals and enabled them to be involved and informed. This
included situations in which carers had been routinely copied into letters and other
documentation, as well as when they had been proactively contacted by staff about
care planning and treatment.
Offers to remain in contact with healthcare professionals and support at follow-up

were highly valued by carers and facilitated opportunities to be involved with the
service user’s recovery process. Carers reflected on the importance of ‘shared
responsibility’ with healthcare services, which helped diminish feelings of isolation
and burden. Feeling supported by services was associated with a perceived
reduction in the carers’ anxiety and burden:
...now I don’t feel so stressed out, because I know that there is so close monitoring of
his progress…That’s a great relief. (HUGHES2011)
Likewise the absence of such support was associated with carers feeling over-
burdened by their caring responsibilities and feeling overlooked by services:
I have almost no communication with the people treating her. I feel as if they are
saying: ‘You’re and outsider, we’re the professionals, you must just stay out of it’.
Nobody tells me how we are supposed to handle this after her discharge. It’s tough not
knowing what I should do if she gets ill. I have a bag full of medicines I’m supposed to
give her. That’s the support apparatus we have. (TRANVAG2008)
Providing clear and comprehensible information

Central to carers’ experience of service were issues relating to individualised
information provision. The findings highlighted the need for healthcare providers to
strike a balance between providing too much information and too little.
Across studies it was also evident that there was a clear need for information
provision to be improved and to be tailored to the specific needs and circumstances
of carers. For example, some reflected on how the timing of the information had an
impact on their understanding and retention of the information provided. Often this
was because of emotional factors that interfered with processing information. This
was particularly noticeable at critical stages in the care pathway, such as during
admission of the service user into acute care or during first episode psychosis:
We were almost in shock when we came here for the first time, we felt as if we were
‘walking beside’ ourselves and could not take it all in. (NORDBY2010)

Providing written information to carers was met with mixed opinions. For some it
allowed information to be revisited regularly and also helped maintain distance
between emotions and information about the disorder:
In a way it’s easier to read about these diseases on a more general level. It does not
seem so personal. I can manage to keep a distance and see it as something many people
suffer from. (NORDBY2010)
However, carers also reflected that the information they received was too
complicated, overwhelming and frightening to read alone. Difficulties such as
dyslexia and language barriers also highlighted the drawbacks of some written
information. Carers suggested that information should be proactively offered,
particularly before a crisis could develop, so that it could be more easily understood
and retained.
Carers were often unaware and unprepared for the challenges that awaited them
over the course of the care pathway. The need for information to be presented earlier
in the process of care was therefore highlighted as crucial in terms of avoiding
distress associated with a lack of information at a later date, particularly at times of
crisis and discharge from acute care:
You discover things gradually after discharge. You do not think to ask of such things
before. (NORDBY2010)
Access to health services

The final theme related to issues around access. Carers suggested that a barrier to
accessing support and services was a lack of knowledge about the structure and
functioning of mental health services. This was perceived to increase levels of stress
and feelings of helplessness in some carers as they reported often not knowing who
to contact in times of crisis. This was particularly evident during first hospital
admission. Carers described needing prompt access to support but instead were
directed from one service to another without clear direction:
I mean one day he had me in tears, I had to walk out of the house and I just walked
into the police station and I spoke to somebody on the desk, and they gave me a little
bit of advice and they told me who to contact and stuff, and the next day I rang, I
actually spoke to somebody but even that was a long process. I phoned them one day
and they said they would get back to me and I said like, I need help now not like
tomorrow or next week. I think they got back to me three months later, it was really
hard to get any kind of help to start with. (JANKOVIC2011)
Carer support groups were considered by some to be a valuable resource in

addressing some of these difficulties as they allowed an opportunity for carers to
access staff who were able to support them in understanding psychiatric services,
how they operate and the sources of help available:

I think for me it was just having a point of contact as well, which I’ve never had
before, I didn’t have any idea of anybody that I could contact or…for any advice or
anything, till I came here. (RILEY2011)
Carers also reported difficulty contacting services when needed. Frustration arose
from the inflexibility of appointments, insufficient scheduling, and a lack of out-of-
hours opening times and availability:
I suppose the major difficulty is when we have crisis …My frustration with them
(Crisis Assessment Treatment team) was their inability to come out one night during
an episode and then another time on a weekend. (MCCANN2011)
In order to improve access to these services carers also highlighted the need for them
to be organised flexibly in terms of timing so as to minimise disruption to caring
responsibilities. The location of services and interventions was also important, for
example having support groups closer to carers’ homes facilitated attendance:
Sometimes their relatives were admitted to places at a distance from their family
home, which caused immense stress for both the carer and service user. (ASKEY2009)
4.2.7 Evidence from qualitative studies of carers’ views and

experiences of interventions for carers
Five studies (LOBBAN2011, MCCANN2011, REID2005, RILEY2011,
WAINWRIGHT) described carers’ experience of interventions and their views on
desirable components of a carer-focused intervention to improve the carers’
experience of care or reduce their burden.
Self-management toolkit
One study provided the views of carers of young people with first episode psychosis
regarding the feasibility of a carer self-management toolkit (LOBBAN2011). Carers
generally welcomed a self-management toolkit aimed at alleviating levels of distress
in carers of people with psychosis. The carers described a number of perceived
benefits, including improved knowledge and understanding as well as reduced
distress and better coping skills. Carers stated that the toolkit should include
information about psychosis, treatment options, and information about the structure
and functioning of mental health services. Information about accessing help during a
crisis and the legal rights of relatives particularly in relations to confidentiality were
particularly important. A modular format was preferred as carers’ felt this would be
more manageable to digest. Carers also encouraged a personalised approach to the
toolkit, which would vary according to the individual’s reading ability. Practical
support in navigating the content was suggested. Carers were emphatic that the
toolkit should supplement and not replace other forms of face-to-face support from
care coordinators and the opportunity to attend important review meetings. The
most appropriate time to receive the toolkit was felt to be after the onset of the
service user’s symptoms but prior to receiving a diagnosis, in order to avoid delays
to treatment.

Group psychoeducation
Three studies examined carers’ views and experiences of group psychoeducation for
carers (RILEY2011, LOBBAN2011, REID2005). Participants expressed positive
feelings about sharing their experiences with other carers. Psychoeducation groups
were considered to provide a safe environment in which carers felt they could speak
freely and be truthful about their relatives’ mental health. The carers felt supported
by each other and by the professionals facilitating the groups. Carers described how
information about the purpose of group psychoeducation needed to be clearer to
allow carers to decide whether it was appropriate for their needs.
Psychoeducation was believed to have a number of practical benefits including a

providing a greater understanding of mental health issues, how to recognise early
warning signs of relapse, and how psychiatric services work. Perceived emotional
benefits included the ability to support other carers in similar circumstances through
involvement as graduate carers, reduced guilt, and improved confidence to deal
with problems resulting in better relationships with the service user. Carers
considered the need for information and advice and the need to hear the stories of
other relatives as particularly important. Carers reported that speaking to others
who had been through similar experiences gave them new ideas about how to cope,
and made them feel less isolated by being able to share and talk openly.
Carers in one study discussed the location and practicalities of delivering a

psychoeducation programme. Several thought that the delivery of the programme
should be delivered in a central location and at different times of the day to give
carers a choice. The majority of carers in this study also stated that home-based
programmes would not be well tolerated as they would disrupt other members of
the family and were unfair for the person hosting the group.
Carer support groups

Four studies described carers’ experience of carer support groups (MCCANN2011,
REID2005, RILEY2011, WAINWRIGHT). Carers reported that these groups
improved their knowledge of mental illness and also helped them to develop better
coping skills. These skills allowed carers to feel more in control and improved their
relationship with the service user. In addition carers gained the skills and knowledge
to be able to proactively access services.
Support groups were valued for addressing the feeling of isolation many carers felt.
The importance of sharing experiences with others carers who were in similar
situations was also preferred over discussing such issues with professionals. The
timing of the group sessions was also important. Because of the positive impact on
improving feelings of isolation and loneliness, carers wanted to be able to access
support groups earlier. Others preferred to attend when they had overcome the
shock of their relative’s illness. Carers also valued the possibility of becoming
graduate carers and helping others going through similar experiences.

A number of barriers to taking part in support groups were highlighted, including
the timing and location of the sessions.
4.2.8 Evidence summary

The thematic synthesis identified five themes that carers of adults with severe
mental illness believed would improve their experience of health and social care
services and reduce carers’ burden. These themes were: (1) building trusting
relationships with healthcare providers; (2) valuing the identity and experience of
the carer; (3) sharing decision making and involvement; (4) providing clear and
comprehensible information; and (5) access to health services. The five major themes
which emerged from the included studies were relevant to all points along the care
pathway. However, some of the themes, for example access to health services or the
provision of clear and compensable information, were also found to be of particular
importance during first episode psychosis and a crisis.
Carers in the included studies also valued carer-focused interventions such as a self-
management toolkit, group psychoeducation and carer support groups as useful
means of receiving information. Group psychoeducation and carer support groups
were also considered to be useful for sharing experiences with others.
4.3 INTERVENTIONS TO IMPROVE CARERS’

EXPERIENCE
4.3.1 Introduction
Definition and aim of review
This aim of this review is to evaluate interventions delivered by health and social
care services to carers of people with severe mental illness, including psychosis and
schizophrenia, to improve their experience of caring. Interventions included in this
review were designed to facilitate the improvement of carers’ experience and reduce
burden. The review aims to evaluate the benefits of the interventions on carer-
focused outcomes and not on the therapeutic outcomes of the service user, thus the
latter were not evaluated or extracted from the papers.
A number of interventions are not included in this review. The provision of financial
and practical support (for example, personal assistance or direct payments) is
outside of the scope of this guideline and is therefore not covered here. Furthermore,
family intervention, which may or may not include the carer or provide carer
outcomes, are evaluated separately in Chapter 9. Interventions where the service
user is included in the majority of sessions are also not included as they are
evaluated in Chapter 9. Additionally, this review does not aim to evaluate the
effectiveness of psychological and pharmacological interventions for carers’ mental
health problems as these are covered by existing NICE guidelines.

Definitions and aim of interventions
Interventions reviewed in this chapter include, but were not limited to, the
following:
Psychoeducation
Psychoeducation/support and education interventions were defined as:
• any structured programme offered individually or in a group involving an
interaction between an information provider and the carer, which has the
primary aim of offering information about the condition, and
• the provision of support and management strategies to carers, and
• delivered to the carer without the service user being present 8.
Where psychoeducation could be either:

• ‘standard’ including only basic information about the nature, prognosis,
symptoms, evolution of illness and treatment of the disorder (including
medication management) and delivered via videos and/information leaflets,
or
• ‘enhanced’ as above but practitioner delivered and including information and
support about additional issues such as how to identify and manage a crisis,
available support services and resources, coping strategies, problem solving,
self-care goals and communication techniques.
Support groups
Support groups were defined as usually a group intervention (although this does not
preclude one-to-one delivery) providing help and support from others. Support
groups can be facilitated by a mental health or social care service provider or a carer
employed by healthcare services (for example, carer support worker). Support
provided is either:
• reciprocal and mutually beneficial for participants who have similar
experiences and who need similar levels of support and (mutual support), or
• primarily in one direction with a clearly defined peer supporter and recipient
of support (peer support).
Self-help interventions
Self-help interventions were defined as:
• including health technologies (for example, written, audio, video and
internet) designed to improve the carers’ experience of care
• including information about the condition and about mental health services
and the support available for the carer
• being guided with support (initial or ongoing support) from a mentor or
healthcare professional, or can be self-directed
• being delivered face-to-face, via telephone or the internet.
8 Psychoeducation involving the service user (with or without the carer) are evaluated in Chapter 7.

4.3.2 Clinical review protocol (interventions to improve carers’
experience)
review protocol can be found in Appendix 6; further information about the search
Table 9: Clinical review protocol summary for the review of interventions to

improve carers’ experience
Review question What modification to health and social services improve the experience of
using services for carers of adults with severe mental illness?
Objectives To evaluate the effectiveness of interventions for improving the experience of
health and social services for carers of people with severe mental illness.
Population Carers of any age who care for adults (18 years of age and over) with severe
mental illness who use health and social services in community settings.
Include papers with a service user population of at least:

66% schizophrenia or
66% schizophrenia + bipolar disorder or
66% schizophrenia + ‘mood disorders’ or
66% undefined severe mental illness
66% bipolar disorder.
Intervention(s) Included interventions
Only interventions delivered directly to carers of people with severe mental
illness will be included. These may include, for example:
• specific interventions for carers
• peer-led interventions for carers (for example, carer support groups)
• changes in the delivery and organisation of services for the benefit of
carers.
Comparison Existing services and alternative strategies
Critical outcomes Carers’:
• quality of life
• mental health (anxiety or depression)
• burden of care (including ‘burnout’, stress, and coping)
• satisfaction with services (validated measures only, specific items will
not be analysed).
Electronic databases Core databases:
CENTRAL, CDSR, DARE, HTA, Embase, MEDLINE, MEDLINE In-Process
Topic specific databases: AEI, ASSIA, BEI, CINAHL, ERIC, IBSS, PsycINFO,
Sociological Abstracts, SSA
Date searched SR: 1995 to June 2013
RCT: database inception to June 2013
Study design Systematic reviews of RCTs
RCT
Review strategy Time-points
• End of intervention
• Up to 6 months’ follow-up (short-term)
• Greater than 6 months’ follow-up (long term)

Where more than one follow-up point within the same period was available,
the latest one was reported.
Analysis
Data were analysed and presented by:
• carer interventions versus any control
• head-to head comparison of carer interventions.
Within these comparisons, subgroups were based on service user diagnosis.
Where data was available, sub-analyses was conducted for UK/Europe

studies.

Twenty four RCTs (N = 1758) met the eligibility criteria for this review: CARRA2007
(Carrà et al., 2007), CHENG2005 (Cheng & Chan, 2005), CHIEN2004A (Chien,
2004a), CHIEN2004B (Chien & Chan, 2004b), CHIEN2007 (Chien & Wong, 2007),
CHIEN2008 (Chien et al., 2008), CHOU2002 (Chou et al., 2002), COZOLINO1988
(Cozolino et al., 1988), GUTIERREZ-MALDONADO2007 (Gutierrez-Maldonado &
Caqueo-Urizar, 2007), KOOLAEE2009 (Koolaee & Etemadi, 2009), LEAVEY2004
(Leavey et al., 2004), LOBBAN2013 (Lobban et al., 2013), MADIGAN2012 (Madigan
et al., 2012), MCCANN2012 (McCann et al., 2012b), PERLICK2010 (Perlick et al.,
2010), POSNOR1992 (Posner et al., 1992), REINARES2004 (Reinares et al., 2004),
SHARIF2012 (Sharif et al., 2012), SMITH1987 (Smith & Birchwood, 1987), SO2006 (So
et al., 2006), SOLOMON1996 (Solomon et al., 1996), SZMUKLER1996 (Szmukler et
al., 1996), SZMUKLER2003 (Szmukler et al., 2003) and VANGENT1991 (Van Gent &
Zwart, 1991). All included studies were published in peer-reviewed journals
between 1987 and 2013. Further information about both included and excluded
studies can be found in Appendix 15a.
Of the 24 eligible trials, 20 (N = 1364) included sufficient data to be included in the

statistical analysis. Three trials did not include any relevant outcomes (CARRA2007,
COZOLINO1988, VANGENT1991) and one trial (N = 225) included critical outcomes
that could not be included in the meta-analyses because of the way the data had
been reported (SOLOMON1996), therefore a brief narrative synthesis is given to
assess whether the findings support or refute the meta-analyses.
The majority of the included trials involved a control arm of treatment as usual
comparing it with psychoeducation (k = 11), a support group (k = 3), a combined
psychoeducation and support group intervention (k = 1), problem-solving
bibliotherapy (k = 1) and self-management (k = 1). Four of the included trials were
three-arm trials comparing two active interventions with treatment as usual. One
trial compared postal psychoeducation with practitioner-delivered standard
psychoeducation, and one trial evaluated group versus individual psychoeducation.

Table 10, Table 11and Table 12 provide an overview of the trials included in each
category. One study (MADIGAN2012) included an arm evaluating an intervention
termed ‘psychotherapy’. However, this arm was not included because the content of
the intervention was poorly described and the suggestion that the intervention was
therapeutic and therefore beyond the scope of this review. Of the eligible trials, 14
included a large proportion (greater than 75%) of service users with a primary
diagnosis of psychosis or schizophrenia and thus the results of sub-analysis are
reported. Only six were based in the UK/Europe and not all trials were included in
the same analysis, thus sub-analysis for UK/Europe based studies was not
conducted.
Table 10: Study information table for trials included in the meta-analysis of carer
interventions versus any control
Psychoeducation versus any Support group versus any

control control
Total no. of trials (k); k = 11; N = 737 k = 3; N = 208
participants (N)
Study ID(s) CHENG2005 CHOU2002
CHIEN2004B CHIEN2004A
CHIEN2007 CHIEN2004B7
GUTIERREZ-MALDONADO2007 CHIEN2008
KOOLAEE20091
LEAVEY2004
MADIGAN2012
POSNOR1992
REINARES2004
SHARIF2012
SO2006
SZMUKLER1996
Country Australia (k = 1) China (k = 4)
Canada (k = 1)
Chile (k = 1)
China (k = 4)
Iran (k = 2)
Ireland (k = 1)
Spain (k = 1)
UK (k = 1)
Year of publication 1992 to 2012 2002 to 2008
Mean age of carers (range) 48.77 years (40.6 to 55.4 years) 2 40.66 years (35.9 to 44.15 years)8
Mean percentage of women 66.38% (31.01 to 100%)3 52.06% (31.01 to 66%)
carers (range)
Mean percentage relationship of Parent = 56.29% Parent = 38.18%
carer to service user Spouse = 19.05% Spouse = 31.56%
Sibling = 6.53% Sibling = 2.85%
(Adult) Child = 6.99% (Adult) Child = 16.51%
Other = 11.14% Other = 10.91%
Mean age of service users (range) 32.88 years (29.1 to 42 years)4 28.52 years (25.35 to 31.68
years) 9
Mean percentage of women 41.77% (27 to 65%)5 46.67% (35.44 to 57.89%)8
service users (range)
Mean percentage of service users 81.82% (0 to 100%)6 100% (100 to 100%)
with primary diagnosis of

psychosis/schizophrenia (range)
Length of treatment (range) 5 to 36 weeks 8 to 24 weeks
Length of follow-up End of treatment only Up to 6 months
CHENG2005 CHOU2002
CHIEN2007 CHIEN2004A
GUTIERREZ-MALDONADO2007 CHIEN2004B
REINARES2004
SO2006 >6 months
CHIEN2004B
Up to 6 months CHIEN2008
CHIEN2004B
KOOLAEE2009
LEAVEY2004
POSNOR1992
SHARIF2012
SZMUKLER1996
>6 months
CHIEN2004B
CHIEN2007
MADIGAN2012
Intervention type Psychoeducation (k = 11) Mutual support (k = 3)
Counselling (psychoeducation + Support group (k = 1)
coping strategies) (k = 1)
Comparisons TAU (k = 8) TAU (k = 3)
Waitlist control (k = 1) Waitlist control (k = 1)
No treatment (k = 2)
Information only (k = 1)
Note. TAU = treatment as usual.
1Two active arms combined.
2 POSNOR1992, LEAVEY2004 and CHENG2005 did not report data.
3 POSNOR1992, SZMUKLER1996, LEAVEY2004 and SHARIF2012 did not report data.
4 LEAVEY2004 and CHENG2005 did not report data.
5 SZMUKLER1996 and CHENG2005 did not report data.
6 100% of service users in REINARES2004 and MADIGAN2012 had a diagnosis of bipolar disorder.
7 CHIEN2004B is a three-arm trial.
8CHOU2002 did not report data.
9 CHOU2002 and CHIEN2004A did not report data.
Table 11: Study information table for trials included in the meta-analysis of carer
interventions versus any alternative management strategy
Psychoeducation + Problem-solving Self-management

support group bibliotherapy versus versus TAU
versus TAU TAU
Total no. of trials (k); k = 1; N = 61 k = 1; N = 124 k = 1; N = 103
participants (N)
Study ID(s) SZMUKLER2003 MCCANN2012 LOBBAN2013
Country UK (k = 1) Australia (k = 1) UK (k = 1)
Year of publication 2003 2012 2013
Mean age of carers 54 years 47.2 years Not reported
Mean percentage of women 82% 82.3% 82.5%
carers
Mean percentage of Parent = 62% Parent = 91.1% Parent = 74%
relationship of carer to service Spouse = 10% Other = 8.9% Other = 26%
user Sibling = 13%

(Adult) Child = 5%
Other = 10%
Mean age of service users Not reported Not reported Not reported
(range)
Mean percentage of women Not reported Not reported Not reported
service users
Mean percentage of service 73% 100% 57%
users with primary diagnosis
of psychosis/ schizophrenia
(range)
Length of treatment 39 weeks 5 weeks 26 weeks
Length of follow-up 7- 12 months Up to 6 months End of treatment only
SZMUKLER2003 MCCANN2012 LOBBAN2013
Intervention type Psychoeducation + Problem-solving Self-management (k
support group (k = bibliotherapy intervention = 1)
1) (k = 1)
Comparisons No treatment (k = TAU (k = 1) TAU (k = 1)
1)
Note. TAU = treatment as usual.
Table 12: Study information table for head-to-head trials comparing different
formats of carer interventions
Enhanced Practitioner-delivered Group

psychoeducation psychoeducation psychoeducation
versus standard versus postal versus individual
psychoeducation psychoeducation psychoeducation
Total no. of trials (k); k = 1; N = 46 k = 1; N = 40 k = 1; N = 225
participants (N)
Study ID(s) PERLICK2010 SMITH1987 SOLOMON1996
Country USA (k = 1) UK (k = 1) USA (k = 1)
Year of publication 2010 1987 1996
Mean age of carers 52.77 years Not reported 55.7 years
Mean percentage of women 84% Not reported 88%
carers
Mean percentage of Parent = 70% Parent = 70% Parent = 76.4%
relationship of carer to service Spouse = 14% Spouse = 17.5% Spouse = 4.4%
user (Adult) child = 14% Other = 12.5% Sibling = 11.1%
Other = 2% (Adult) child = 5.8%
Other = 2.2%
Mean age of service users 34.72 years 36.4 years 35.8 years
Mean percentage of women 63% 22% Not reported
service users
Mean percentage of service 0%1 100% 63.5%
users with primary diagnosis
of psychosis/ schizophrenia
Length of treatment 12 to 15 weeks 4 weeks 10 weeks
Length of follow-up End of treatment only Up to 6 months 7- 12 months
PERLICK2010 SMITH1987 SOLOMON1996
Intervention type Enhanced psycho- Practitioner delivered Group psycho-
education (k = 1) psychoeducation (k = 1) education (k = 1)
Comparisons Standard psycho- Postal psychoeducation Individual psycho-
education (k = 1) (k = 1) education (k = 1)
Note. 1 100% of service users had a diagnosis of bipolar disorder.

4.3.4 Clinical evidence for any intervention versus any control
In the included trials, the interventions were compared with a variety of control
groups that were categorised as any control (treatment as usual, attention control,
waitlist control and no treatment). Further information about the control group used
in each trial can be found in Table 10, Table 11and Table 12.
Psychoeducation versus control

Evidence from each important outcome and overall quality of evidence are
presented in
. The full evidence profiles and associated forest plots can be found in Appendix 17
and Appendix 16, respectively.
Low to very low quality evidence from up to seven studies (N = 399), showed that
psychoeducation was more effective than control in improving carers’ experience of
care and these effects are maintained at long-term follow-up. No difference was
observed between groups in quality of life or satisfaction with services. Although no
difference was observed between groups in psychological effect at the end of the
intervention and at short-term follow-up, one study (N = 18) provided high quality
evidence that psychoeducation is more effective than control at long-term follow-up.
Support group versus control

presented in Table 14. The full evidence profiles and associated forest plots can be
found in Appendix 17 and Appendix 16, respectively.
Low to very low quality evidence from up to three studies (N = 194) showed that
support groups improved the experience of caring at the end of the intervention and
at short-term follow-up but no benefit was observed at long-term follow-up. One
study with 70 participants presented low quality evidence that support groups were
more effective than control for reducing psychological distress at the end of the
intervention and at short-term follow-up.
Psychoeducation plus support group versus control

One study with 49 participants found no difference between psychoeducation plus

support group and control in terms of the experience of caring and psychological
distress. No other follow-up data or other critical outcome data were available.

Table 13: Summary of findings table for psychoeducation compared with any
control
Patient or population: Carers of adults with severe mental illness

Intervention: Psychoeducation
Comparison: Any control
Outcomes Illustrative comparative risks* (95% CI) No. of Quality
Corresponding risk participants of the
Psychoeducation (studies) evidence
(GRADE)
Experience of Mean experience of caring (end of intervention) in the 399 ⊕⊝⊝⊝
caring - intervention groups was 1.03 standard deviations higher (7 studies) very low1,2
end of intervention (0.36 to 1.69 higher)
Experience of Mean experience of caring (up to 6 months’ follow-up) in 215 ⊕⊝⊝⊝
caring - up to 6 the intervention groups was 0.92 standard deviations higher (4 studies) very low1,2
months’ follow-up (0.32 to 1.51 higher)
Experience of Mean experience of caring (>6 months’ follow-up) in the 151 ⊕⊝⊝⊝
caring - >6 intervention groups was 1.29 standard deviations higher (3 studies) very low1,2
months’ follow-up (0.18 to 2.4 higher)
Quality of life - end Mean quality of life (end of intervention) in the intervention 41 ⊕⊕⊝⊝
of intervention groups was 0.31 standard deviations higher (0.31 lower to (1 study) low1,3
0.93 higher)
Satisfaction with Mean satisfaction with services (end of intervention) in the 39 ⊕⊕⊝⊝
services - end of intervention groups was 0.42 standard deviations higher (1 study) low1,3
intervention (0.22 lower to 1.06 higher)
Satisfaction with Mean satisfaction with services (up to 6 months’ follow-up) 39 ⊕⊕⊝⊝
services - up to 6 in the intervention groups was 0.41 standard deviations (1 study) low1,3
months’ follow-up higher (0.23 lower to 1.04 higher)
Psychological Mean psychological distress (end of intervention) in the 86 ⊕⊝⊝⊝
distress - end of intervention groups was 0.3 standard deviations lower (0.84 (2 studies) very
intervention lower to 0.24 higher) low1,2,3
Psychological Mean psychological distress (up to 6 months’ follow-up) in 86 ⊕⊕⊝⊝
distress- up to 6 the intervention groups was 0.34 standard deviations lower (2 studies) low1,3
months’ follow-up (0.76 lower to 0.08 higher)
Psychological Mean psychological distress (> 6 months’ follow-up) in the 18 ⊕⊕⊕⊕
distress - >6 intervention groups was 1.79 standard deviations lower (1 study) high
months’ follow-up (3.01 to 0.56 lower)
Note. CI = confidence interval.
*The basis for the assumed risk (for example, the median control group risk across studies) is provided in the
footnotes below. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group
and the relative effect of the intervention (and its 95% CI).
1 Concerns regarding risk of bias.
2 Concerns regarding heterogeneity.
3 CI crosses clinical decision threshold (SMD of 0.2 or -0.2; RR of 0.75 or 1.75).

Table 14: Summary of findings table for support group compared with any control

Intervention: Support groups
Outcomes Illustrative comparative risks* (95% CI) No. of Quality of
Corresponding risk participants the
Support groups (studies) evidence
(GRADE)
Experience of caring - Mean experience of caring (end of intervention) in 194 ⊕⊝⊝⊝
end of intervention the intervention groups was 1.16 standard (3 studies) very low1,2,3
deviations higher (0.36 to 1.96 higher)
Experience of caring - Mean experience of caring (up to 6 months’ 166 ⊕⊕⊝⊝
up to 6 months’ follow- follow-up) in the intervention groups was 0.67 (3 studies) low1,3
up standard deviations higher (0.35 to 0.99 higher)
Experience of caring - Mean experience of caring (>6 months’ follow-up) 123 ⊕⊝⊝⊝
>6 months’ follow-up in the intervention groups was 1.95 standard (2 studies) very low1,2,3,4
deviations lower
(4.22 lower to 0.31 higher)
Psychological distress - Mean psychological distress (end of intervention) 70 ⊕⊕⊝⊝
end of intervention in the intervention groups was 0.99 standard (1 study) low1,3
deviations lower (1.48 to 0.49 lower)
Psychological distress - Mean psychological distress (up to 6 months’ 70 ⊕⊕⊝⊝
up to 6 months’ follow- follow-up) in the intervention groups was 0.99 (1 study) low1,3
up standard deviations lower (1.48 to 0.49 lower)
Note. CI = confidence interval
2 Concerns regarding heterogeneity.
3 Studies all based in East Asia - may not be applicable to UK setting.
4 Confidence interval crosses clinical decision threshold.
Table 15: Summary of findings table for psychoeducation plus support group
compared with any control

Intervention: Psychoeducation + support group
Outcomes Illustrative comparative risks* (95% CI) No. of participants Quality of
Corresponding risk (studies) the
Psychoeducation + support group evidence
(GRADE)
Experience Mean experience of caring (>6 months’ follow-up) in the 49 ⊕⊕⊝⊝
of caring - intervention groups was 0.05 standard deviations higher (0.51 (1 study) low1,2
>6 months’ lower to 0.61 higher)
follow-up
*The basis for the assumed risk (for example, the median control group risk across studies) is provided in the footnotes
below. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative
effect of the intervention (and its 95% CI).
2 Confidence interval crosses decision making threshold.

Self-management versus control
One study with 86 participants found no difference between groups in terms of

experience of caring and psychological distress at the end of the intervention.
Table 16: Summary of findings table for self-management compared with any
control

Intervention: Self-management
Self-management evidence
(GRADE)
Experience of Mean experience of caring (end of intervention) in the 86 ⊕⊕⊕⊝
caring - end of intervention groups was 0.19 standard deviations lower (1 study) moderate1
Psychological Mean psychological distress (end of intervention) in the 86 ⊕⊕⊕⊝
distress - end intervention groups was 0.32 standard deviations lower (1 study) moderate1
of intervention (0.73 lower to 0.09 higher)
*The basis for the assumed risk (for example, the median control group risk across studies) is provided in the footnote
below. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative
effect of the intervention (and its 95% CI).
Problem-solving bibliotherapy versus control

One study with 114 participants found no difference between groups in terms of the
experience of caring. The same study provided low quality evidence that problem-
solving bibliotherapy was effective at improving quality of life at short-term follow-
up (although no difference was observed at the end of the intervention).

Table 17: Summary of findings table for problem-solving bibliotherapy compared
with any control

Intervention: Problem-solving bibliotherapy
Outcomes Illustrative comparative risks* (95% CI) No. of Quality
Corresponding risk participants of the
Problem-solving bibliotherapy (studies) evidence
(GRADE)
Experience of caring – Mean experience of caring (end of 114 ⊕⊕⊝⊝
end of intervention intervention) in the intervention groups was (1 study) low1,2
0.17 standard deviations higher (2.11 lower to
2.45 higher)
Experience of caring - Mean experience of caring (up to 6 months’ 114 ⊕⊕⊝⊝
up to 6 months’ follow-up) in the intervention groups was (1 study) low1,2
follow-up 1.09 standard deviations higher (0.34 lower to
2.52 higher)
Quality of life - end of Mean quality of life (end of intervention) in 114 ⊕⊕⊝⊝
intervention the intervention groups was 0.14 standard (1 study) low1,2
deviations higher (0.23 lower to 0.5 higher)
Quality of life - up to 6 Mean quality of life (up to 6 months’ follow- 114 ⊕⊕⊝⊝
months’ follow-up up) in the intervention groups was 0.5 (1 study) low1,2
standard deviations higher 0.12 to 0.87
higher)
Psychological distress Mean psychological distress (end of 114 ⊕⊕⊕⊝
– intervention) in the intervention groups was (1 study) moderate1
end of intervention 1.57 standard deviations lower (1.79 to 1.35
lower)
Psychological distress- Mean psychological distress (up to 6 months’ 111 ⊕⊕⊕⊝
up to 6 months’ follow-up) in the intervention groups was (1 study) moderate1
follow-up 1.54 standard deviations lower (1.95 to 1.13
lower)
footnotes below. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison
group and the relative effect of the intervention (and its 95% CI).
2 CI crosses clinical decision making threshold
Enhanced psychoeducation versus standard psychoeducation


One trial with 43 participants provided moderate quality evidence that enhanced
psychoeducation was more effective than standard psychoeducation in improving
experience of caring and self-care behaviour when measured at the end of the
intervention. No difference was observed between groups in carer mental health. No
follow-up data were available.
Practitioner-delivered versus postal-delivered standard psychoeducation

One study with 40 participants provided data for this comparison. There was no
evidence of a difference between groups in family burden and psychological distress
at the end of the intervention and up to 6 months’ follow-up. No other follow-up
data or other critical outcome data were available.

Table 18: Summary of findings table for enhanced psychoeducation compared
with standard psychoeducation

Intervention: Enhanced psychoeducation
Comparison: Standard psychoeducation
Enhanced psychoeducation evidence
(GRADE)
Experience of Mean experience of caring (end of intervention) in 43 ⊕⊕⊕⊝
caring - end of the intervention groups was 0.64 standard (1 study) moderate1
intervention deviations higher (0.3to 1.25 higher)
Carer mental Mean carer mental health (end of intervention) in 43 ⊕⊕⊕⊝
health - end of the intervention groups was 0.32 standard (1 study) moderate1
intervention deviations higher (0.29 lower to 0.92 higher)
Self-care - end Mean self-care (end of intervention) in the 43 ⊕⊕⊕⊝
of intervention intervention groups was 0.68 standard deviations (1 study) moderate1
lower (1.31 to 0.06 lower)
in the footnote below. The corresponding risk (and its 95% CI) is based on the assumed risk in the
comparison group and the relative effect of the intervention (and its 95% CI).
Table 19: Summary of findings table for practitioner-delivered compared with

postal-delivered standard psychoeducation

Intervention: Psychoeducation-practitioner delivered
Comparison: Psychoeducation-postal delivered
Standard psychoeducation (practitioner-delivered) evidence
(GRADE)
Family burden – Mean family burden (end of intervention) in the 40 ⊕⊕⊝⊝
end of intervention groups was 0.41 standard deviations lower (1 study) low1,2
Family burden - Mean family burden (- up to 6 months’ follow-up) in 40 ⊕⊕⊝⊝
up to 6 months’ the intervention groups was (1 study) low1,2
follow-up 0.41 standard deviations lower (1.03 lower to 0.22
higher)
Psychological Mean psychological distress (end of intervention) in the 40 ⊕⊕⊝⊝
distress - end of intervention groups was (1 study) low1,2
intervention 0.38 standard deviations lower (1 lower to 0.25 higher)
Psychological Mean psychological distress (up to 6 months’ follow- 40 ⊕⊕⊝⊝
distress - up to 6 up) in the intervention groups was 0 standard (1 study) low1,2
months’ follow- deviations higher (0.62 lower to 0.61 higher)
up
*The basis for the assumed risk (for example, the median control group risk across studies) is provided in the footnotes.
The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of

the intervention (and its 95% CI).
Individual versus group enhanced psychoeducation versus treatment as

usual
The trial eligible for this review (SOLOMON1996) could not be included in meta-
analysis. The study reported no significant difference between groups in terms of
carers’ burden or satisfaction with services.
4.3.5 Clinical evidence summary

The limited evidence suggests that psychoeducation is effective in reducing carers’
burden and these effects are maintained at long-term follow-up. Furthermore,
evidence suggests that although no immediate benefit can be found at the end of the
intervention, psychoeducation can reduce psychological distress in the long term.
Support groups were also found to be effective in improving carers’ experience of
caring and reducing psychological distress. However, these findings should be
viewed with caution as the studies included in this review are based in East Asia
and the services provided there are not directly comparable to the UK. In addition,
there was limited evidence that enhanced psychoeducation (providing information,
as well as focusing on self-carer skills, coping skills and problem-solving) was more
effective than standard psychoeducation (information only) in improving the
experience of caring and self-care behaviour at the end of the intervention. However,
longer-term effects are not known. Self-management was not found to be beneficial
over control on any critical outcomes. However, this was based on a single high
quality study and a trend favouring self-management was observed. Problem-
solving bibliotherapy was not found to be effective at improving any critical
outcomes at the end of the intervention, however, it was found to improve quality of
life at short-term follow-up. Finally, there was no detectable difference in
effectiveness between psychoeducation delivered by post or delivered by a
practitioner, or between group and individual psychoeducation.
4.4 HEALTH ECONOMICS EVIDENCE

No studies assessing the cost effectiveness of interventions aiming to improve carers’
experience of caring and of health and social care services were identified by the
systematic search of the economic literature undertaken for this guideline. Details on
the methods used for the systematic search of the economic literature are described
in Chapter 3.
The clinical studies on interventions, mainly psychoeducation, aiming to improve

carers’ experience of caring and of health and social care services included in the
guideline systematic literature review (GUTIERREZ-MALDONADO2007,
SHARIF2012, CHENG2005, SZMUKLER1996) described interventions consisting of
13 sessions on average (range 6 to 26). These programmes are usually delivered by
either a psychologist or psychiatric nurse/psychiatrist to an average group of seven
people (range 1 to 9) and have an average duration of 1.5 hours (range 1 to 2). The

unit cost of a clinical psychologist is £136 per hour of client contact in 2011/12 prices
(Curtis, 2012). This estimate has been based on the median full-time equivalent basic
salary for Agenda for Change salaries band 8a of the April 2012 NHS Staff Earnings
Estimates (Health and Social Care Information Centre, 2012). It includes basic salary,
salary oncosts, travel, overheads and capital overheads, but does not take into
account qualification costs because the latter are not available for clinical
psychologists. The unit cost of a mental health nurse is £76 per hour of client contact
in 2011/12 prices (Curtis, 2012). This estimate has been based on the median full-
time equivalent basic salary for Agenda for Change salaries band 5 of the April-June
2012 NHS Staff Earnings Estimates for Qualified Nurses (Health and Social Care
Information Centre, 2012). It includes basic salary, salary oncosts, qualifications,
overheads and capital overheads, and travel. The unit cost of a psychiatric
consultant is £289 per hour of client contact in 2011/12 prices (Curtis, 2012). This
estimate has been based on the Electronic Staff Records system that shows the mean
full-time equivalent total earnings for a psychiatric consultant in April to June 2012
(Health and Social Care Information Centre, 2012). It includes basic salary, salary
oncosts, qualifications, ongoing training, overheads and capital overheads. Based on
the estimated resource utilisation associated with interventions aiming to improve
carers’ experience of caring and of services (as described above) and the unit cost of
a clinical psychologist, a mental health nurse and a psychiatric consultant the
average cost per person participating in such a programme would range between
£190 and £1,095 (mean of £582) in 2011/12 prices.

Relative value placed on the outcomes considered:
The main aim of the qualitative review was to evaluate carers’ experience of health
and social care services. The outcomes of interest were any themes and specific
issues that carers identified as improving or diminishing their experience of health
and social care. Furthermore, the GDG aimed to evaluate the effectiveness of
interventions designed to improve the carers’ experience of caring. The outcomes the
GDG considered to be critical for carers were their:
• quality of life
• mental health (anxiety or depression)
• burden of care (including ‘burnout’, stress and coping)
• satisfaction with services
Trade-off between clinical benefits and harms

The factors identified by the qualitative review revealed a broad range of issues that
resonated with the experience of the carers, service users and healthcare professional
members of the GDG.
The qualitative analysis revealed that carers thought a key determinant of their
experience of services and experience of caring was building trusting relationships
with healthcare professionals. An empathic and understanding healthcare

professional allows the carer to build confidence in their role as a carer and reduces
feelings of stress and burden.
Two linked themes were identified in the qualitative literature. Carers felt that
services should identify and value their experience and involve them in decision
making. This theme also included issues about confidentiality—carers felt that
confidentiality was often used as a reason to exclude them from receiving important
information about the service user’s care and treatment, resulting in a stressful,
burdensome and isolated experience for them. This theme was prevalent throughout
the care pathway and specifically during first episode psychosis, crises and
subsequent exacerbations, as well as during the planning of discharge from a
hospital. The GDG used these findings to make recommendations about the
involvement of carers and the negotiation of information sharing among the service
user, the carer and the healthcare professionals. Furthermore, in taking a broad
overview of all the themes identified, combined with the collective experience of the
whole GDG, the GDG came to the view that the guideline should explicitly support
collaboration among the carer, service user and healthcare professional through all
phases of care, where this is possible, while respecting the independence of the
service user.
Importantly, a theme affecting both carers and service users is access to services.
Carers expressed a need to have easy access to services, interventions and support
for the service user, which thus reduces the carer’s own burden and stress. Carers
discussed the importance of swift access to reliable services at all points in the care
pathway but particularly during a crisis and during first episode psychosis. Carers
stated that other practical concerns such as flexible services in terms of times and
dates, and appropriate location of services also reduced carers’ burden and stress.
Furthermore, carers also stressed the need for access to support for themselves.
Carer support groups were said to be of great value as an informal way of receiving
regular support from others who have had similar experiences.
Carers valued the provision of clear and comprehensible information. However

what was also evident from the literature was that carers valued the information
more at certain points in the care pathway. For example, carers stated they needed
more information during the early stages of assessment and first episode psychosis,
but the information should not be too copious (and thus overwhelming) or too brief
(and therefore of little use). Furthermore, carers stressed that an individualised
approach to providing information should be used and that the information given to
them should be in a format and delivered at times tailored to the specific needs of
the carer and the service user.
A key point present across identified themes was that carers, like service users,
would like an atmosphere of optimism and hope when in contact with services and
healthcare professionals. The GDG considered this important and decided to reflect
this in the recommendations.

Carers were generally positive about, and suggested components for, a self-
management toolkit. They were concerned, however, that healthcare professionals
might see the toolkit as a reason to disengage with them. Carers’ experience of group
psychoeducation was positive overall, but carers stated that the aim of the group
should be very clear in order to avoid disappointment if the group did not meet
individual needs. Carer support groups were found to be very useful and valued by
carers.
The literature evaluating the effectiveness of the carer-focused interventions was

limited but promising. Psychoeducation and support groups both provided evidence
of benefits on carers’ experience of care, quality of life and satisfaction. A self-
management toolkit and bibliotherapy intervention did not statistically show any
benefit over control, although a trend favouring the interventions was observed. The
review of carer-focused interventions included trials of people with psychosis,
schizophrenia or bipolar disorder as well as mixed diagnosis populations. Although
the majority of the available evidence was with a psychosis and schizophrenia
population, the GDG believed that the issues faced by carers of adults with
psychosis and schizophrenia would be applicable to carers of adults with bipolar
disorder or other severe mental illnesses. The analyses were highly underpowered
and the GDG considered that the further trials would increase the power of the
analysis and could show a benefit over control.
On the basis of the quantitative review of interventions for carers, the GDG decided
that interventions specifically aimed to help carers should be provided. The evidence
did not permit a recommendation of a particular type of intervention. However, it
was evident, from both the qualitative and quantitative literature, that carers require
support, education and information and therefore the GDG made a recommendation
that states the components of an intervention that should be provided for the carer.
Trade-off between net health benefits and resource use

No economic studies assessing the cost effectiveness of interventions aimed at
improving carers’ experience were identified. The cost of providing such
interventions was estimated at roughly between £190 and £1,095 (mean of £582) in
2011/12 prices. The GDG judged this cost to be small taking into account the effects
of the intervention, leading to a reduction in carers’ burden, potential depression
and other health vulnerabilities which may be costly to other parts of the NHS,
especially considering that the burden of care can last for many years and increase
carer morbidity and stress. In addition, increased knowledge and improved
confidence helps carers to contribute to care more effectively. Despite the small,
emerging evidence base, interventions that aim to improve carers’ experience of
caring and of services were judged by the GDG to represent good value for money
and be worth the investment.
Quality of the evidence

The evidence ranged from very low to moderate quality across critical outcomes.
Reasons for downgrading included: risk of bias in the included studies and high

heterogeneity or lack of precision in confidence intervals. Wide confidence intervals
were also a major concern when evaluating the evidence. However, although
variance was observed in the effect size across studies, the direction of effect was
consistent across most and the small number of participants in the included trials
could have contributed to the lack of precision. Furthermore, some of the included
studies for support groups were based in settings that may not be appropriate to the
UK healthcare setting (for example, East Asia). In these instances, the evidence was
downgraded for indirectness. The evidence showed a benefit of support groups for
the carer, but the GDG was cautious about making a recommendation specifically
for support groups for this reason. However, the GDG believed that there was also
qualitative evidence of great benefits of support groups and therefore could still be
considered when drafting recommendations.
Other considerations
At the time of drafting the 2014 guideline, the Service User Experience in Adult Mental
Health guidance was in the public domain. The GDG judged that it was of prime
importance that a cross-reference to this guidance was made because the 2014
guideline has not re-reviewed any of the qualitative evidence for service user
experience.
The GDG considered all identified themes to be important and as a basis for
recommendations. However, they also discussed that the recommendations should
not be biased towards the carer over the service user’s needs, but should be
complementary. This is likely to benefit both the carer and the service user because a
carer who is well informed and supported is more likely to provide better support
and care for the service user. This is also important because carers are an integral
part of family intervention. The GDG considered that although this chapter does not
explicitly review family intervention (the evidence for it was reviewed for the 2009
guideline [see Chapter 9] ), it remains essential that the offer of any carer-focused
intervention is a part of family intervention. Consideration should be given to the
most appropriate timing for psychoeducation offered on an individual basis.
The GDG discussed the term ‘psychoeducation’ used to describe some of the
interventions reviewed. The GDG felt that the term was outdated and that it does
not reflect the nature of current interventions, which do not aim to ‘teach’ things.
Interventions that showed some benefit for the carer usually included aspects that
also provided emotional support for the carer. The GDG decided to use the term
‘education and support’, which they judged to be appropriate in underlining the
dyadic relationship between the healthcare professional or worker providing the
education and support and the carer to emphasise the fact that the intervention is
usually more than the provision of written information. The GDG also decided that
the recommendation should contain guidance about what education and support
programmes should entail.

4.6 RECOMMENDATIONS
4.6.1 Clinical practice recommendations
4.6.1.1 Offer carers of people with psychosis or schizophrenia an assessment
(provided by mental health services) of their own needs and discuss with
them their strengths and views. Develop a care plan to address any
identified needs, give a copy to the carer and their GP and ensure it is
reviewed annually. [new 2014]
4.6.1.2 Advise carers about their statutory right to a formal carer’s assessment
provided by social care services and explain how to access this. [new 2014]
4.6.1.3 Give carers written and verbal information in an accessible format about:
• diagnosis and management of psychosis and schizophrenia
• positive outcomes and recovery
• types of support for carers
• role of teams and services
• getting help in a crisis.
When providing information, offer the carer support if necessary. [new 2014]
4.6.1.4 As early as possible negotiate with service users and carers about how
information about the service user will be shared. When discussing rights to
confidentiality, emphasise the importance of sharing information about risks
and the need for carers to understand the service user’s perspective. Foster a
collaborative approach that supports both service users and carers, and
respects their individual needs and interdependence. [new 2014]
4.6.1.5 Review regularly how information is shared, especially if there are
communication and collaboration difficulties between the service user and
carer. [new 2014]
4.6.1.6 Include carers in decision-making if the service user agrees. [new 2014]
4.6.1.7 Offer a carer-focused education and support programme, which may be part
of a family intervention for psychosis and schizophrenia, as early as possible
to all carers. The intervention should:
• be available as needed
• have a positive message about recovery. [new 2014]
4.6.2 Research recommendation

4.6.2.1 What are the benefits for service users and carers for family intervention
combined with a carer-focused intervention compared with family
intervention alone?[ new 2014]

5 PREVENTING PSYCHOSIS AND
SCHIZOPHRENIA: TREATMENT OF
AT RISK MENTAL STATES
This chapter is new for the 2014 guideline. It is taken from a review undertaken for
Psychosis and Schizophrenia in Children and Young People (NCCMH, 2013 [full
guideline]) of recognition of at risk mental states and of pharmacological,
psychosocial and dietary interventions for people at risk of developing psychosis
and schizophrenia. The review of the interventions was updated by a subsequent
systematic review by Stafford and colleagues (2013). The populations in the studies
in the review included people over the age of 18 years and were, therefore, deemed
relevant by the GDG for the 2014 guideline.
5.1 INTRODUCTION
Over the past 2 decades there has been a wealth of research examining the
possibility of early recognition of psychosis, with an emphasis on reducing duration
of untreated psychosis (DUP), which has been shown to be associated with poor
outcomes. More recently, there has also been increased interest in the identification
of people who are at high risk of developing a first psychotic episode with the hope
that intervention could prevent or delay the development of a psychosis. Many
people who go on to develop a psychosis experience a variety of psychological,
behavioural and perceptual disturbances prior to the psychosis, sometimes for
several months. Previously described as a prodromal period, most studies have
adopted other terms including at risk, or ultra-high risk, states.
5.1.1 Recognition, identification and treatment strategies for at risk

mental states
Recent studies have examined the feasibility of detecting and treating people in the
‘at risk’ stage, prior to the development of psychosis. This approach rests on three
assumptions: (1) it is possible to detect such people; (2) these people will be at
markedly increased risk of later psychosis; and (3) an effective intervention will
reduce this risk. There is evidence to support (1) and (2) in people with a strong
family history of psychosis who are therefore at high genetic risk (Miller et al., 2001)
and in those reporting particular perceptual abnormalities (Klosterkotter et al., 2001).
When those at risk have been identified, there is the question of what can effectively
be done to prevent, delay or ameliorate psychosis. To date, there have been nine
RCTs, each using similar operational definitions of ’at risk’, which have reported
findings regarding antipsychotic medication, omega-3 polyunsaturated fatty acids
and/or psychological interventions including CBT. These studies have been
conducted in Australia (McGorry et al., 2002; Phillips et al., 2009), North America
(Addington et al., 2011; McGlashan et al., 2006) and Europe (Amminger et al., 2010;
Bechdolf et al., 2012; Morrison et al., 2007; Morrison et al., 2004) and have aimed to

achieve one or more of the following outcomes: to prevent, delay or ameliorate rates
of transition to psychosis; to reduce severity of psychotic symptoms; to reduce
distress and emotional dysfunction; and to improve quality of life.
The following therapeutic approaches have been identified:
• pharmacological interventions:
- olanzapine
- risperidone
• dietary interventions:
- omega-3 fatty acids
• psychological interventions:
- cognitive behavioural therapy (CBT)
- integrated psychological therapy
- supportive counselling.
Some researchers have combined more than one intervention in order to improve the
likelihood of achieving the intended outcomes. For example, an antipsychotic
medication can be combined with a psychological therapy such as cognitive therapy,
or several psychosocial interventions may be combined (such as cognitive therapy,
CRT and family intervention). These combinations do not form a homogenous group
and therefore cannot be analysed together in a meta-analysis.
5.2 CLINICAL REVIEW PROTOCOL FOR AT RISK

MENTAL STATES FOR PSYCHOSIS AND
SCHIZOPHRENIA
A summary of the review protocol, including the review questions, information
about the databases searched and the eligibility criteria used for this section of the
guideline can be found in Table 20. (A full review protocol can be found in
Appendix 6 and further information about the search strategy can be found in
Appendix 13).
Table 20: Clinical review protocol for the review of at risk mental states for
Review questions For people who are at risk of developing psychosis1 and schizophrenia (at risk
mental state), does the provision of pharmacological, psychological or
psychosocial and/or dietary interventions improve outcomes? 2
Objectives To evaluate if pharmacological, psychological or psychosocial and/or dietary
interventions improve outcomes for people who are at risk of developing
psychosisand schizophrenia.
Population Inclusion: People considered to be at high risk of developing a first episode
psychosis.
Exclusion: Study samples consisting of individuals with a formal diagnosis of

psychosis, schizophrenia or bipolar disorder.
Interventions Licensed antipsychotics drugs.2

Psychological interventions, including:
• CBT
• CRT
• Counselling and supportive psychotherapy
• Family intervention (including family therapy)
• Psychodynamic psychotherapy and psychoanalysis
• Psychoeducation
• Social skills training
• Arts therapies
Dietary interventions, including:

• Any dietary/nutritional supplements
Comparison Alternative management strategies:
• Placebo
• Treatment as usual
• Waitlist
Any of the above interventions offered as an alternative management strategy.
Criticaloutcomes • Transition to psychosis.
• Time to transition to psychosis.
Important but not • Mental state (symptoms, depression, anxiety, mania)
criticaloutcomes • Mortality (including suicide)
• Global state
• Psychosocial functioning
• Social functioning
• Leaving the study early for any reason
• Adverse effects (including effects on metabolism, EPS, hormonal changes
and cardiotoxicity)
Electronic databases Core databases: Embase, MEDLINE, MEDLINE In-Process.
Topic-specific databases: PsycINFO.
Date searched 2011 to October 2013
Study design Systematic reviews
Review strategy • This updates an existing review (Stafford et al., 2013) in which searches for
systematic reviews and RCTs were conducted to November 2011. RCT
evidence was identified from the Stafford review (2013), and from searches
conducted for Chapter 5 of CG155, generated to May 2012.
• Two independent reviewers reviewed the full texts obtained through sifting
all initial hits for their eligibility according to the inclusion criteria outlined
in this protocol.
• The initial approach was to conduct a meta-analysis evaluating the benefits
and harms of pharmacological, psychological, dietary and combination
treatment. However, in the absence of adequate data, the literature was
presented via a narrative synthesis of the available evidence.
• Unpublished data was included when the evidence was accompanied by a
trial report containing sufficient detail to properly assess the quality of the
data. The evidence had to be submitted with the understanding that data
from the study and a summary of the study’s characteristics would be
published in the full guideline. Unpublished data was not included
wherethe evidence submitted was commercial and in confidence.
Note. People who are at risk of developing psychosis and those who have early psychosis but do not have a
1
formal diagnosis of either schizophrenia or bipolar disorder.

5.2.1 Ethical considerations
There has been considerable debate within the scientific and clinical communities
regarding the desirability of ‘labelling’ people as being at high risk of developing
psychosis and schizophrenia. This is partly because the rates of transition suggest
that the majority of such samples (between 80 and 90%) do not convert to first
episode psychosis within a 12-month period (that is, there are many ‘false positives’),
and there is some evidence that these rates are declining (Yung et al., 2007). This may
mean exposing people to risks associated with the label, such as unnecessary stigma
(Bentall & Morrison, 2002; Yang et al., 2010), restrictions that people may impose
upon themselves (such as avoidance of stress) (Warner, 2001) and unwanted
consequences for employment or obtaining insurance, for example (Corcoran et al.,
2010). There are also concerns about the risks of exposure to unnecessary treatments
with potential adverse effects within this population, and hence the risks and
benefits of any intervention must be balanced carefully (Bentall & Morrison, 2002;
Warner, 2001). The proposal to include a psychosis risk syndrome, so-called
‘attenuated psychotic disorder’ in DSM-5, has led to many concerns for such reasons
(Carpenter, 2009; Corcoran et al., 2010; Morrison et al., 2010). Nevertheless, the GDG
considered that the benefits for individuals, families and the wider society that could
result from preventing the development of psychosis is so substantial, given the
often devastating effects that many people experience as a result of psychosis, that a
full review of strategies to prevent psychosis in at risk states outweighed these
important ethical considerations.
5.3 PHARMACOLOGICAL INTERVENTIONS

5.3.1 Studies considered
The GDG selected an existing review (Stafford et al., 2013) as the basis for this
section of the guideline. The existing Stafford review (2013) included four RCTs (N =
358 ) providing relevant clinical evidence and meeting the eligibility criteria for the
review: MCGLASHAN2003 (McGlashan et al., 2003), MCGORRY2002 (McGorry et
al., 2002), PHILLIPS2009 (Phillips et al., 2009), RUHRMANN2007 (Ruhrmann et al.,
2007). Three studies were published in peer reviewed journals between 2002 and
2007 and one study contained unpublished data (PHILLIPS2009). All studies
contained participants who were judged to be at risk of developing psychosis on the
basis of a clinical assessment identifying prodromal features. Further information
about both included and excluded studies can be found in (Stafford et al., 2013).
Of the four included trials, there was one comparing olanzapine with placebo, two
comparing risperidone plus CBT with supportive counselling, one comparing
risperidone plus CBT with placebo plus CBT, and one comparing amisulpride and a
needs based intervention with the needs based intervention alone. PHILLIPS2009
had three treatment groups and was included in two of the pair wise comparisons
(see Table 21 for a summary of the study characteristics).

Table 21: Study information table for trials of antipsychotic medication
Olanzapine Risperidone + Risperidone + Amisulpride + NBI

versus CBT versus CBT versus versus NBI
placebo supportive placebo + CBT
counselling
Total no. of 1 (N = 60) 2 (N = 130) 1 (N = 87) 1 (N = 124)
studies (N)
Study ID MCGLASHA (1) MCGORRY2002 PHILLIPS2009 RUHRMANN2007
N2003 (2) PHILLIPS2009
Screening tool SIPS1 (1) Not reported CAARMS2 ERIraos4
(2) CAARMS2
Diagnosis At-risk mental Ultra-high risk Ultra-high risk
state mental state mental state
Mean age 17.8 (range 12 (1) 20 (range 14 to 17.9 (not reported)3 25.6 (not reported)
(range) to 36) 28)
(2) 17.9 (not
reported)3
Sex (% male) 65 (1) 58 393 56
(2) 393
Ethnicity (% 67 (1)–(2) Not reported Not reported Not reported
white)
Mean (range) 8 (range 5 to (1) 1.3 (range 1 to 2) 2 (not reported) 118.7 (range 50 to 800)
medication dose 15) (2) 2 (not reported)
(mg/day)
Sessions of N/A (1) Mean (SD) Up to 35 hours Not reported
therapy sessions attended:
CBT: 11.3 (8.4);
Supportive
counselling: 5.9
(4.3).
(2) Up to of 35 hours
of CBT or
supporting
counselling
Treatment 52 (1) 26 52 12
length (weeks) (2) 52
Treatment 104 (1) 156 to 208 104 N/A
follow-up (2) 104
(weeks)
Setting Specialist (1)–(2) Specialist Specialist Specialist clinic/ward
clinic/ward clinic/ward clinic/ward
Country US (1)–(2) Australia Australia Germany
Note. N = Total number of participants. CBT= Cognitive behavioural therapy; NBI=Needs based intervention
1 Structured Interview for Prodromal Symptoms.
2 Comprehensive assessment of at-risk mental states.
3 In whole study (N = 115; PHILLIPS2009 is a three way comparison evaluating risperidone, CBT and SC).
4 Early Recognition Inventory
Psychosis and schizophrenia in adults
106
5.3.2 Clinical evidence for olanzapine versus placebo
Efficacy
One study (N = 60) compared olanzapine with placebo. At 1 year post-treatment 16
participants had transitioned to psychosis and there was no statistically significant
difference between groups. Effects on symptoms of psychosis, depression, and
mania were also not significant. Evidence from each reported outcome and overall
quality of evidence are presented in Table 22 and Table 23.
Side effects
There were more olanzapine dropouts at 1 year, but the difference was not
statistically significant. Participants taking olanzapine gained significantly more
weight at 1-year post-treatment. Furthermore, compared with the placebo group the
sitting pulse of participants in the olanzapine group increased significantly more
from baseline to post-treatment (very low quality evidence). Effects on standing
pulse were not significant. At 104 weeks’ follow-up transition to psychosis and side
effects were measured, however, the data were considered unusable because there
were fewer than 10 people remaining in each group. Evidence from each reported
outcome and overall quality of evidence are presented in Table 22 and Table 23.
5.3.3 Clinical evidence for risperidone plus CBT versus supportive

counselling
Efficacy
Two studies (N = 130) compared risperidone plus CBT with supportive counselling.
Within the first 26 weeks of treatment, fewer people receiving risperidone plus CBT
transitioned to psychosis (defined as the development of a DSM-IV psychotic
disorder), but these trials included 17 events (very low quality evidence). By 52
weeks’ follow-up the effect was no longer significant and this remained non-
significant at 156 to 208 weeks’ follow-up. At follow-up, only data for completers
were reported and therefore a sensitivity analysis for transition to psychosis was
conducted, assuming dropouts had made transition. In sensitivity analysis the effect
remained non-significant. Both studies reported mean endpoint scores for symptoms
of psychosis, quality of life, depression, anxiety, mania and psychosocial
functioning. No significant differences between treatment groups were found on
these outcomes at post-treatment or follow-up. At post-treatment, there was no
dropout in one study (MCGORRY2002) and dropout in the other (PHILLIPS2009)
was similar between groups. Evidence from each reported outcome and overall
quality of evidence are presented in Table 24, Table 25, and Table 26.
Side effects
For the participants for whom side effect data were reported, there was no
significant difference between groups at post-treatment (see Table 24).

Table 22: Summary of findings table for outcomes reported for olanzapine versus placebo at 52 weeks post-treatment
Outcome or subgroup Study ID Number of Effect estimate Heterogeneity Quality of

studies / (SMD or RR) evidence
participants [95% CI] (GRADE)a
Total symptoms (SMD) MCGLASHAN2003 K = 1, N = 59 -0.12 [-0.63, 0.39] N/A Very low1,2,3
Positive symptoms (SMD) MCGLASHAN2003 K = 1, N = 59 -0.40 [-0.91, 0.12] N/A Very low1,2,3
Negative symptoms (SMD) MCGLASHAN2003 K = 1, N = 59 0.05 [-0.46, 0.56] N/A Very low1,2,3
Global state (severity) (SMD) MCGLASHAN2003 K = 1, N = 59 -0.17 [-0.68, 0.34] N/A Very low1,2,3
Depression (SMD) MCGLASHAN2003 K = 1, N = 59 0.32 [-0.19, 0.83] N/A Very low1,2,3

Mania (SMD) MCGLASHAN2003 K = 1, N = 59 -0.15 [-0.66, 0.36] N/A Very low1,2,3
Psychosocial functioning (SMD) MCGLASHAN2003 K = 1, N = 59 -0.16 [-0.67, 0.35] N/A Very low1,2,3
Transition to psychosis (RR) MCGLASHAN2003 K = 1, N = 60 0.43 [0.17, 1.08] N/A Very low1,2,3
Leaving the study early for any reason MCGLASHAN2003 K = 1, N = 60 1.59 [ 0.88, 2.88] N/A Very low1,2,3
(RR)
Weight gain (kg; SMD) MCGLASHAN2003 K = 1, N = 59 1.18 [0.62, 1.73]* N/A Very low1,2,3
Sitting pulse (beats per minute MCGLASHAN2003 K = 1, N = 60 0.61 [0.08, 1.13]* N/A Very low1,2,3
[BPM]; SMD)
Standing pulse (BPM; SMD) MCGLASHAN2003 K = 1, N = 59 0.37 [-0.15, 0.88] N/A Very low1,2,3
Note. aThe GRADE approach was used to grade the quality of evidence for each outcome.
*Favours placebo
1 Serious risk of bias (including unclear sequence generation and allocation concealment and missing data)
2 Optimal information size (for dichotomous outcomes, OIS = 300 events; for continuous outcomes, OIS = 400 participants) not met
3 Serious risk of reporting bias

Table 23: Summary of findings table for outcomes reported for olanzapine versus placebo at 104 weeks’ follow-up (change
scores from post-treatment until follow-up when no treatment was received)
Outcome or subgroup Study ID Number of Effect estimate (SMD or RR) Heterogeneity Quality of
studies/ [95% CI] evidence
participants (GRADE)a
Leaving the study early for any MCGLASHAN2003 K = 1, N = 60 0.98 [0.71, 1.35] N/A Very low1,2,3
reason (RR)
Note. aThe GRADE approach was used to grade the quality of evidence for each outcome.1Serious risk of bias (including unclear sequence generation
and allocation concealment and missing data)
3Serious risk of reporting bias

Table 24: Summary of findings table for outcomes reported for risperidone plus CBT versus supportive counselling at
post-treatment

studies / (SMD or RR) evidence
Total symptoms (SMD) MCGORRY2002 K = 2, N = 102 0.15 [-0.39, 0.70] (P = 0.12); I² = 59% Very low1,2,3
PHILLIPS2009
Positive symptoms (SMD) MCGORRY2002 K = 2, N = 130 0.02 (-0.33, 0.37) (P = 0.39); I² = 0% Very low1,2,3
PHILLIPS2009
Negative symptoms (SMD) MCGORRY2002 K = 2, N = 130 0.13 (-0.68, 0.94) (P = 0.02); I² = 81% Very low1,2,3
PHILLIPS2009
Depression (SMD) MCGORRY2002 K = 2, N = 130 0.24 (-0.12, 0.59) (P=0.003) I² = 88% Very low1,2,3
PHILLIPS2009
Mania (SMD) MCGORRY2002 K = 1, N = 59 -0.20 [-0.71, 0.32] N/A Very low1,2,3
Anxiety (SMD) MCGORRY2002 K = 1, N = 59 -0.15 [-0.66, 0.36] N/A Very low1,2,3
Psychosocial functioning (SMD) PHILLIPS2009 K = 1, N = 43 -0.12 [-0.73, 0.49] N/A Very low1,2,3
Quality of life (SMD) MCGORRY2002 K = 2, N = 130 -0.13 [-0.49, 0.22] (P = 0.31); I² = 2% Very low1,2,3
PHILLIPS2009
Transition to psychosis (RR) MCGORRY2002 K = 2, N = 130 0.35 [0.13, 0.95] (P = 0.44); I² = 0% Very low1,2,3
PHILLIPS2009
Leaving the study early for any reason MCGORRY2002 K = 2, N = 130 0.76 [0.28, 2.03] N/A [no events Very low1,2,3
(RR) PHILLIPS2009 observed by
MCGORRY2002]
EPS (RR) PHILLIPS2009 K = 1, N = 21 0.55 [0.13, 2.38] N/A Very low1,2,3
Note.
aThe GRADE approach was used to grade the quality of evidence for each outcome.
1Serious risk of bias (including unclear sequence generation, allocation concealment, raters unblind to psychological intervention, trial registration not
found, uneven sample sizes and missing data)

Table 25: Summary of findings table for outcomes reported for risperidone plus CBT versus supportive counselling at 52
weeks’ follow-up
Outcome or subgroup Study ID Number of studies Effect estimate Heterogeneity Quality of

/ participants (SMD or RR) evidence
[95% CI] (GRADE)a
Total symptoms (SMD) MCGORRY2002 K=2, N = 101 0.07 [-0.32, 0.46] (P = 0.39); I² = 0% Very low1,2,3
PHILLIPS2009
Positive symptoms (SMD) MCGORRY2002 K=2, N = 101 0.05 [-0.35, 0.44] (P = 0.90); I² = 0% Very low1,2,3
PHILLIPS2009
Negative symptoms (SMD) MCGORRY2002 K=2, N = 101 0.08 [-0.31, 0.47] (P = 0.41); I² = 0% Very low1,2,3
PHILLIPS2009
Depression (SMD) MCGORRY2002 K=2, N = 68 0.15 [-0.33, 0.62] (P = 0.93); I² = 0% Very low1,2,3
PHILLIPS2009
Mania (SMD) MCGORRY2002 K=1, N = 59 0.00 [-0.51, 0.51] N/A Very low1,2,3
Anxiety (SMD) MCGORRY2002 K = 1, N = 59 0.06 [-0.45, 0.57] N/A Very low1,2,3
Psychosocial functioning (SMD) MCGORRY2002 K = 1, N = 59 0.00 [-0.51, 0.51] N/A Very low1,2,3
Quality of life (SMD) MCGORRY2002 K=2, N = 102 -0.07 [-0.46, 0.32] (P = 0.84); I² = 0% Very low1,2,3
PHILLIPS2009
Transition to psychosis (RR) MCGORRY2002 K = 2, N = 130 0.63 [0.33, 1.21] (P = 0.61); I² = 0% Very low1,2,3
PHILLIPS2009
Leaving the study early for any reason MCGORRY2002 K=2, N = 130 0.85 [0.43, 1.67] (P = 0.19); I² = 43% Very low1,2,3
(RR) PHILLIPS2009
1Serious risk of bias (including unclear sequence generation, allocation concealment, raters unblind to psychological intervention, trial registration could
not be found and missing data).

3Serious risk of reporting bias.

Table 26: Summary of findings table for outcomes reported for risperidone plus CBT versus supportive at 156 to 208
weeks’ follow-up
Outcome or subgroup Study ID Number of studies Effect estimate Heterogeneity Quality of

/ participants (SMD or RR) evidence
[95% CI] (GRADE)a
Total symptoms (SMD) MCGORRY2002 K = 1, N = 41 -0.33 [-0.96, 0.29] N/A Very low1,2,3
Positive symptoms (SMD) MCGORRY2002 K = 1, N = 41 -0.04 [-0.66, 0.58] N/A Very low1,2,3
Negative symptoms (SMD) MCGORRY2002 K = 1, N = 41 -0.24 [-0.87, 0.38] N/A Very low1,2,3
Depression (SMD) MCGORRY2002 K = 1, N = 41 0.23 [-0.39, 0.86] N/A Very low1,2,3
Mania (SMD) MCGORRY2002 K = 1, N = 41 -0.36 [-0.98, 0.27] N/A Very low1,2,3
Anxiety (SMD) MCGORRY2002 K = 1, N = 41 0.14 [-0.49, 0.76] N/A Very low1,2,3
Psychosocial functioning (SMD) MCGORRY2002 K = 1, N = 41 -0.15 [-0.77, 0.47] N/A Very low1,2,3
Quality of life (SMD) MCGORRY2002 K = 1, N = 41 0.08 [-0.54, 0.71] N/A Very low1,2,3
Completer analysis: transition to MCGORRY2002 K = 1, N = 41 0.59 [0.34, 1.04] N/A Very low1,2,3
psychosis (RR)
Number of participants requiring MCGORRY2002 K = 1, N = 41 0.51 [0.19, 1.33] N/A Very low1,2,3
hospitalisation (RR)
Leaving the study early for any MCGORRY2002 K = 1, N = 59 0.57 [0.26, 1.28] N/A Very low1,2,3
reason (RR)
1Serious risk of bias (including unclear sequence generation, allocation concealment, raters unblind to psychological intervention, trial registration
could not be found and missing data)


5.3.4 Clinical evidence for risperidone plus CBT versus placebo plus
CBT
Efficacy
One study (N = 87) compared risperidone plus CBT with placebo plus CBT. By 52
weeks post-treatment, seven participants in each group had transitioned to
psychosis (defined as the development of a DSM-IV psychotic disorder) and there
was no significant difference between groups. Differences in symptoms of psychosis,
depression, psychosocial functioning and quality of life were not significant, and
dropout was similar between groups. Evidence from each reported outcome and
overall quality of evidence are presented in Table 27.
Side effects
For participants whom side effect data were reported experienced EPS (as measured
by the UKU Neurologic Subscale). However, there was no significant difference
between groups. Evidence from each reported outcome and overall quality of
evidence are presented in Table 27.
5.3.5 Clinical evidence for amisulpride plus a ‘needs based

intervention’ versus a ‘needs based intervention’
Efficacy
One study (N = 102) compared amisulpride and a needs based intervention with the
needs based intervention alone. Transition to psychosis was not reported. Within six
months, effects on total and negative symptoms of psychosis were not significant,
but amisulpride was associated with a moderate reduction in positive symptoms,
and depression. Evidence from each reported outcome and overall quality of
evidence are presented in Table 28.
Side effects
The addition of amisulpride was associated with a moderate reduction in dropout.
Of the 19 participants who dropped out of the amisulpride group, three were a
result of adverse events provoked by prolactin-associated symptoms, that is,
galactorrhoea in two participants and sexual dysfunction in another. There was
however no significant difference between groups at post treatment. Evidence from
each reported outcome and overall quality of evidence are presented in Table 28.

Table 27: Summary evidence profile for outcomes reported for risperidone plus CBT versus placebo plus CBT at 52 weeks
post-treatment
Outcome or subgroup Study ID Number of studies/ Effect estimate Heterogeneity Quality of

participants (SMD or RR) [95% evidence
CI] (GRADE)a
Total symptoms (SMD) PHILLIPS2009 K = 1, N = 51 -0.24 [-0.79, 0.31] N/A Very low1,2,3
Positive symptoms (SMD) PHILLIPS2009 K = 1, N = 51 -0.07 [-0.62, 0.48] N/A Very low1,2,3
Negative symptoms (SMD) PHILLIPS2009 K = 1, N = 51 0.12 [-0.43, 0.67] N/A Very low1,2,3
Psychosocial functioning (SMD) PHILLIPS2009 K = 1, N = 9 0.24 [-0.31, 0.78] N/A Very low1,2,3
Quality of life (SMD) PHILLIPS2009 K = 1, N = 52 -0.23 [-0.78, 0.33] N/A Very low1,2,3
Transition to psychosis (RR) PHILLIPS2009 K = 1, N = 51 1.02 [0.39, 2.67] N/A Very low1,2,3
Leaving the study early for any reason (RR) PHILLIPS2009 K = 1, N = 56 1.09 [0.62, 1.92] N/A Very low1,2,3
EPS (RR) PHILLIPS2009 K = 1, N = 87 0.87 [0.18, 4.24] N/A Very low1,2,3
1Serious risk of bias (including unclear sequence generation, allocation concealment, trial registration not found, uneven sample sizes).

Table 28: Summary evidence profile for outcomes reported for amisulpride plus a ‘needs-based intervention’ versus a ‘needs-
based intervention’ at up to 6 months’ follow-up
Outcome or subgroup Study ID Number of studies / Effect estimate Heterogeneity Quality of

participants (SMD or RR)[95% evidence
CI] (GRADE)a
Total symptoms (SMD) RUHRMANN2007 K = 1, N = 102 -0.36 [-0.75, 0.04] N/A Very low1,2,3
Positive symptoms (SMD) RUHRMANN2007 K = 1, N = 102 -0.53 [-0.93, -0.13] N/A Very low1,2,3
Negative symptoms (SMD) RUHRMANN2007 K = 1, N = 102 -0.26 [-0.65, 0.14] N/A Very low1,2,3
Depression (SMD) RUHRMANN2007 K = 1, N = 102 -0.51 [-0.91, -0.11] N/A Very low1,2,3
Leaving the study early for any RUHRMANN2007 K = 1, N = 124 0.59 [0.38, 0.94] N/A Very low1,2,3
reason (RR)
Leaving the study early due to side RUHRMANN2007 K = 1, N = 124 6.36 [0.34, 120.67] N/A Very low1,2,3
effects (RR)
Note. a The GRADE approach was used to grade the quality of evidence for each outcome.
1Serious risk of bias (including unclear sequence generation, allocation concealment, raters unblind to psychological intervention, trial registration could
not be found and missing data)


5.3.6 Clinical evidence summary for pharmacological interventions
Four RCTs (N = 358) conducted in people with an at-risk mental state for psychosis
or schizophrenia were reviewed. One study investigated the effect of an
antipsychotic medication alone against placebo (two studies investigated the effect
of an antipsychotic medication in combination with CBT against a psychological
therapy and one study investigated the effect of antipsychotic medication in
combination with a needs based intervention against a needs based intervention
alone. The findings suggest that antipsychotic medication is no more effective than a
psychological intervention or placebo in preventing transition to psychosis and has
little or no effect in reducing psychotic symptoms. What is more, olanzapine
treatment can result in significant weight gain.
5.4 DIETARY INTERVENTIONS

section of the guideline. The existing Stafford review (2013) included one RCT (N =
81) providing relevant clinical evidence that met the eligibility criteria for this
review: AMMINGER2010 (Amminger et al., 2010) (see Table 29 for a summary of the
study characteristics).
Table 29: Study information table for trials of dietary interventions
Omega-3 fatty acids versus placebo

Total no. of studies (N) 1 (N = 81)
Study ID AMMINGER2010
Screening tool Positive and Negative Syndrome Scale
(PANSS)
Diagnosis Ultra-high risk mental state
Mean age (range) 16.4 (not reported)
Sex (% male) 33
Ethnicity (% white) Not reported
Mean (range) medication dose (mg/day) 1200
Treatment length (weeks) 12
Treatment follow-up (weeks) 52
Setting Specialist clinic/ward
Country Austria
Funding Stanley Medical Research Institute
5.4.2 Clinical evidence for omega-3 fatty acids versus placebo

One study compared omega-3 polyunsaturated fatty acids (ω-3 PUFAs) with
placebo. At 12 weeks post-treatment significantly more participants in the placebo
group had transitioned to psychosis (defined as the development of a DSM-IV
psychotic disorder). However, there were only nine events in total. As only data for
completers were reported a sensitivity analysis for transition to psychosis was

conducted, assuming dropouts had made transition, and the effect became non-
significant. No other outcomes were reported at this time point. At 52 weeks’ follow-
up including all participants randomised the effect was significant. Large effects on
total symptoms of psychosis, positive and negative symptoms of psychosis,
depression and psychosocial functioning also favoured omega-3 fatty acids at 52
weeks’ follow-up. Dropout after 52 weeks was low and similar between groups.
Evidence from each reported outcome and overall quality of evidence are presented
in Table 30 and Table 31.
5.4.3 Clinical evidence summary for dietary interventions

One RCT (N = 81) comparing omega-3 fatty acids with placebo was reviewed.
Although the study was well conducted, sample sizes were small. The findings
suggest that omega-3 fatty acids may be effective at preventing transition to
psychosis and improving symptoms of psychosis, depression and psychosocial
functioning in young people (low quality evidence). However, owing to the paucity
of evidence (lack of independent replication) no robust conclusions can be made.

Table 30: Summary of findings table for outcomes reported for omega-3 fatty acids versus placebo at 12 weeks post-treatment

studies/ (SMD or RR) evidence
Completer analysis: transition to psychosis (RR) AMMINGER2010 K = 1, N = 76 0.13 [0.02, 0.95]* N/A Low2, 3
*Favours omega-3 fatty acids
Table 31: Summary of findings table for outcomes reported for omega-3 fatty acids versus placebo at 52 weeks’ follow-up
Effect estimate Quality of

Number of studies
Outcome or subgroup Study ID (SMD or RR) [95% Heterogeneity evidence
/ participants
CI] (GRADE)a
Total symptoms (SMD) AMMINGER2010 K = 1, N = 81 -1.26 [-1.74, -0.78]* N/A Low 1, 2
Positive symptoms (SMD) AMMINGER2010 K = 1, N = 81 -2.08 [-2.63, -1.54]* N/A Low1, 2
Negative symptoms (SMD) AMMINGER2010 K = 1, N = 81 -2.22 [-2.77, -1.66]* N/A Low 1, 23

Depression (SMD) AMMINGER2010 K = 1, N = 81 -0.56 [-1.01, -0.12]* N/A Low21, 2
Psychosocial functioning (SMD) AMMINGER2010 K = 1, N = 81 -1.28 [-1.76, -0.80]* N/A Low1, 2
Transition to psychosis (RR) AMMINGER2010 K = 1, N = 81 0.18 [0.04, 0.75]* N/A Low1, 2

Leaving the study early for any reason (RR) AMMINGER2010 K = 1, N = 81 1.46 (0.26 to 8.30) N/A Low1, 2
Note. The GRADE approach was used to grade the quality of evidence for each outcome.
a
*Favours omega-3 fatty acids

5.5 PSYCHOSOCIAL INTERVENTIONS
section of the guideline. The existing Stafford review (2013) included seven RCTs (N
= 879 ) providing relevant clinical evidence met the eligibility criteria for this review:
ADDINGTON2011 (Addington et al., 2011), MORRISON2004 (Morrison et al., 2004),
MORRISON2011 (Brown et al., 2011), PHILLIPS2009 (Phillips et al., 2009),
VANDERGAAG2012 (Attux et al., 2013). Of these, two contained some unpublished
data (MORRISON2004 and PHILLIPS2009) and the remaining trials were published
between 2004 and 2012. Further information about the included and excluded
studies can be found in Stafford et al. (2013).
Of the seven included trials, five studies compared individual CBT with supportive
counselling, one study compared a multimodal intervention (integrated
psychological therapy) with supportive counselling, and one study compared a
similar multimodal intervention with standard care (see Table 32 for a summary of
the study characteristics).
5.5.2 Clinical evidence for CBT versus supportive counselling

Five RCTs (N = 672) compared CBT with supportive counselling. Within the first 26
weeks of treatment CBT did not significantly reduce transition to psychosis (defined
as the development of a DSM-IV psychotic disorder) compared with supportive
counselling, observing 40 events in total (N = 591). However, at 52 weeks’ follow-up,
moderate quality evidence found a medium effect of CBT on transition to psychosis.
As one study in the meta-analysis only reported data for completers a sensitivity
analysis for transition to psychosis (assuming dropouts had made transition) was
conducted. In sensitivity analysis this effect remained significant. Furthermore, at 78
weeks’ (or more) follow-up CBT was significantly associated with fewer transitions
to psychosis; however, this did not remain significant in sensitivity analysis.
Combined effects for total symptoms of psychosis, positive and negative symptoms
of psychosis, depression, anxiety, psychosocial functioning and quality of life were
not significant at any time point. However, one study (VANDERGAAG2012)
reported secondary outcomes only for participants who had not transitioned;
participants with the most severe symptoms were omitted from these analyses. In
sensitivity analyses excluding this study, there was a significant effect for positive
symptoms at 52 weeks’ follow-up, but effects for other outcomes remained non-
significant. Dropout was similar between groups within the first 6 months. Evidence
from each reported outcome and overall quality of evidence are presented in Table
33, Table 34, and Table 35.

Table 32: Study information table for trials of psychosocial interventions
CBT versus supportive Integrated psychological therapy Integrated psychological therapy

counselling versus supportive counselling versus standard care
Total no. of studies 5 (N = 672) 1 (N = 128) 1 (N = 79)
(N)
Study ID (1) ADDINGTON2011 BECHDOLF2012 NORDONTOFT2006
(2) MORRISON2004
(3) MORRISON2011
(4) PHILLIPS2009
(5) VANDERGAAG2012
Screening tool (1) SIPS Early Recognition Inventory and Interview ICD-10
(2) PANSS for the Retrospective Assessment of the
(3)-(5) CAARMS Onset of Schizophrenia
Diagnosis ‘At risk/ultra-high risk mental Early initial prodromal state Schizotypal disorder
state’
Mean age (range) (1) 20.9 (not reported) 25.8 (not reported) (2) 24.9 (not reported)
(2) 22 (range 16 to 36)
(3) 20.7 (range 14 to 34)
(4) 17.9 (not reported)1
(5) 22.7
Sex (% male) (1) 71 66 67
(2) 67
(3) 63
(4) 391
(5) 49
Ethnicity (1) 57 Not reported Not reported
(% white) (2) Not reported
(3) 88
(4)-(5) Not reported
Sessions of therapy (1) CBT and supportive 25 individual therapy sessions; 15 group Needs based
counselling: up to 20 sessions; 12 CRT sessions; three
(2) CBT: 26; supportive information and counselling of relatives

counselling: 13 sessions
(3) CBT: 26; supportive
counselling: not reported
(4) Up to of 35 hours
(5) CBT: up to 26; supportive
counselling: not reported
Treatment length (1) 26 52 104
(weeks) (2) 52
(3) 26
(4) 52
(5) 26
Treatment follow-up (1) 78 104 N/ A
(weeks) (2) 156
(3) 104
(4) 52
(5) 78
Setting (1) Specialist clinic/ward Specialist clinic/ward Specialist clinic/ward
(2)-(3) Not reported
(4) Specialist clinic/ward
(5) Mental health centres
(multisite)
Country (1) Canada Germany Denmark
(2)-(3) UK
(4) Australia
(5) Netherlands
Note. 1In the whole study (a three-way comparison evaluating risperidone, CBT and supportive counselling, N = 115).

Table 33: Summary of findings table for outcomes reported for CBT versus supportive counselling at post-treatment (within
26 weeks)
Outcome or Study ID Number of Effect estimate Heterogeneity Quality of evidence

subgroup studies/ (SMD or RR) [95% (GRADE)a
participants CI]
Total symptoms (SMD) ADDINGTON2011 K = 2, N = 123 0.004[-0.32, 0.40] (P = 0.77); I² = 0% Low1,2
PHILLIPS2009
Completer analysis: ADDINGTON2011 K = 4, N = 489 -0.12 [-0.30, 0.06] (P = 0.90); I² = 0% Moderate1
positive symptoms MORRISON2011
(SMD) PHILLIPS2009
VANDERGAAG2012
Negative symptoms ADDINGTON2011 K = 2, N = 123 0.17 [-0.19, 0.53] (P = 0.54); I² = 0% Low1,2
(SMD) PHILLIPS2009
Depression (completer ADDINGTON2011 K = 4, N = 478 0.12 [-0.20, 0.47] (P = 0.03); I² = 67% Low1,2
analysis) (SMD) MORRISON2011
PHILLIPS2009
VANDERGAAG2012
Anxiety (social; SMD) MORRISON2011 K = 1, N = 172 0.01 [-0.28, 0.31] N/A Low1,2
Psychosocial functioning ADDINGTON2011 K = 3, N = 291 0.02 [-0.22, 0.26] (P = 0.96); I² = 0% Low1,2
(SMD) MORRISON2011
PHILLIPS2009
Quality of life (completer MORRISON2011 K = 3, N = 383 0.01 [-0.19, 0.21] (P = 0.78); I² = 0% Low1,2
analysis) (SMD) PHILLIPS2009
VANDERGAAG2012
Transition to psychosis ADDINGTON2011* K = 4, N = 591 0.62 [0.29, 1.31] (P = 0.31); I² = 17% Low1,2
(completer analysis) (RR) MORRISON2011
PHILLIPS2009
VANDERGAAG2012
Leaving the study early ADDINGTON2011 K = 3, N = 411 1.01 [0.75, 1.36] (P = 0.93); I² = 0% Low1,3
for any reason (RR) MORRISON2011
PHILLIPS2009
Note. aThe GRADE approach was used to grade the quality of evidence for each outcome. bThe sensitivity analysis excluded VANDERGAAG2012* 15
weeks during treatment 1Serious risk of bias (including unclear sequence generation, trial registration could not be found, missing data).
2 Optimal information size (for dichotomous outcomes, OIS = 300 events; for continuous outcomes, OIS = 400 participants) not met. 3 I2 ≥ 50%, p<.05

Table 34: Summary of findings table for outcomes reported for CBT versus supportive counselling at 52 weeks’ follow-up
Outcome or subgroup Quality of

Number of Effect estimate
evidence
Study ID studies/ (SMD or RR) Heterogeneity
(GRADE)
participants [95% CI] a
Total symptoms (SMD) ADDINGTON2011 K = 3, N = 154 0.05 [-0.27, -0.37] (P = 0.08); I² = 0% Low1,2
MORRISON2004
PHILLIPS2009
Positive symptoms (completer analysis) (SMD) ADDINGTON2011 K = 5, N = 493 -0.17 [-0.35, 0.01] (P = 0.47); I² = 0% Moderate1,
MORRISON2004
MORRISON2011
PHILLIPS2009
VANDERGAAG2012
Negative symptoms (SMD) ADDINGTON2011 K = 3, N = 154 0.11 [-0.21, 0.43] (P = 0.95); I² = 0% Low1,2
MORRISON2004
PHILLIPS2009
Completer analysis: depression (SMD) ADDINGTON2011 K = 3, N = 385 -0.05 [-0.25, 0.15] (P = 0.63); I² = 0% Low1,2
MORRISON2011
VANDERGAAG2012
Anxiety (social; SMD) MORRISON2011 K = 1, N = 188 0.15 [-0.15, 0.44] N/A Low1,2
Psychosocial functioning (SMD) ADDINGTON2011 K = 2, N = 240 -0.10 [-0.36, 0.15] (P = 0.70); I² = 0% Low1,2
MORRISON2011
Completer analysis: quality of life (SMD) MORRISON2011 K = 3, N = 329 -0.01[-0.23, 0.21] (P = 0.75); I² = 0% Low1,2
PHILLIPS2009
VANDERGAAG2012
Completer analysis: transition to psychosis ADDINGTON2011 K = 5, N = 645 0.54 [ 0.34, 0.86] (P = 0.64); I² = 0% Moderate2
(RR) MORRISON2004
MORRISON2011
PHILLIPS2009
VANDERGAAG2012

Leaving the study early for any reason (RR) ADDINGTON2011 K = 5, N = 665 1.03 [0.82, 1.30] (P = 0.83); I² = 0% Low1,2
MORRISON2004
MORRISON2011
PHILLIPS2009
VANDERGAAG2012
bThe sensitivity analysis excluded VANDERGAAG2012
*Favours CBT
1Serious risk of bias (including unclear sequence generation, , trial registration could not be found, missing data).
Table 35: Summary of findings table for outcomes reported for CBT versus supportive counselling ≥78 weeks’ follow-up

studies/ (SMD or RR) evidence
Total symptoms (SMD) ADDINGTON2011 K = 1, N = 51 -0.04 [-0.59, 0.51] N/A Low1,2
Completer analysis: positive symptoms (SMD) ADDINGTON2011 K = 3, N = 256 -0.17 [-0.42, 0.07] (P = 0.72); I² = 0% Low1,2
MORRISON2011
VANDERGAAG2012
Sensitivity analysis: positive symptoms (SMD)b ADDINGTON2011 K = 2, N = 116 -0.14 [-0.50, 0.23] (P = 0.45); I² = 0% -
MORRISON2011
Negative symptoms (SMD) ADDINGTON2011 K = 1, N = 51 -0.10 [-0.65, 0.45] N/A Low1,2

Completer analysis: depression (SMD) ADDINGTON2011 K = 3, N = 352 -0.11[-0.36, 0.13] (P = 0.49); I² = % Low1,2
MORRISON2011
VANDERGAAG2012
Sensitivity analysis: depression (SMD)b ADDINGTON2011 K = 2, N = 112 -0.05[-0.46, 0.37] (P = 0.27); I² = 19% -
MORRISON2011
Anxiety (social; SMD) MORRISON2011 K = 1, N = 58 -0.46 [-0.99, 0.06] N/A Low1,2

Psychosocial functioning (SMD) ADDINGTON2011 K = 2, N = 116 -0.03 [-0.45, 0.40] (P = 0.25); I² = 25% Low1,2

MORRISON2011
Completer analysis: quality of life (SMD) MORRISON2011 K = 2, N = 188 0.18 [-0.10, 0.47] (P = 0.39); I² = 0% Low1,2
VANDERGAAG2012
Sensitivity analysis: quality of life (SMD)b MORRISON2011 K = 1, N = 48 0.40[-0.17, 0.98] N/A -
Completer analysis: transition to psychosis (RR) ADDINGTON2011 K = 4, N = 570 0.63 [0.40, 0.99] (P = 0.48); I² = 0% Low1,2
MORRISON2011
MORRISON2004
VANDERGAAG2012
Sensitivity analysis: transition to psychosis ADDINGTON2011 K = 4, N = 595 0.55 [0.25, 1.19] (P = 0.002); I² = Low1,2
(assuming dropouts transitioned; RR) MORRISON2011 79%
MORRISON2004
VANDERGAAG2012
Leaving the study early for any reason (RR) ADDINGTON2011 K = 4, N = 593 1.09 [0.88, 1.35] (P = 0.58); I² = 0% Low1,2
MORRISON2004
MORRISON2011
VANDERGAAG2012
bThe sensitivity analysis excluded VANDERGAAG2012

1Seriousrisk of bias (including unclear sequence generation, , trial registration could not be found, missing data).

5.5.3 Clinical evidence for integrated psychological therapy versus
supportive counselling
One study (N = 128) compared integrated psychological therapy with supportive
counselling in participants in the early initial prodromal state. Integrated
psychological therapy included individual CBT, group skills training, CRT and
family treatments, in the absence of antipsychotic medication. Transition to
psychosis was defined as either the development of attenuated (subclinical) or
transient symptoms (subthreshold psychosis) or a DSM-IV psychotic disorder. At 1-
year post-treatment fewer people receiving integrated psychological therapy
transitioned. The effect was maintained at 2 years’ follow-up. Dropout was similar
between groups at 1 year and 2 years post-treatment. Other symptoms were not
reported as outcomes, although the PANSS and Global Assessment of Functioning
(GAF) were recorded at baseline. Evidence from each reported outcome and overall
quality of evidence are presented in Table 36 and Table 37.
5.5.4 Clinical evidence for integrated psychological therapy versus

standard care
One study (N = 79) compared integrated psychological therapy with standard care
in first contact patients diagnosed with schizotypal disorder. Within 12 months,
fewer people receiving integrated psychotherapy transitioned to psychosis, but the
effect was not quite significant after 24 months. There was no effect for positive or
negative symptoms of psychosis at either time point. Dropout was similar between
groups at 12 months and 24 months. Evidence from each reported outcome and
overall quality of evidence are presented in Table 38 and Table 39.

Table 36: Summary of findings table for outcomes reported for integrated psychological therapy versus
supportive counselling at 52 weeks post-treatment
Outcome or subgroup Study ID Number of studies Effect estimate (SMD Heterogeneity Quality of evidence
/ participants or RR) [95% CI] (GRADE)a
Transition to psychosis (RR) BECHDOLF2012 K = 1, N = 125 0.19 [0.04, 0.81]* N/A Very low1,2,3
Leaving the study early for any reason (RR) BECHDOLF2012 K = 1, N = 128 1.55 [0.68, 3.53] N/A Very low1,2,4
Note.aThe GRADE approach was used to grade the quality of evidence for each outcome.
*Favours integrated psychological therapy
1 Serious risk of bias (missing data).
3Serious risk of indirectness (participants classified as in the early initial prodromal state as opposed to a high risk mental state and transition is defined as the
development of either attenuated/transient symptoms or a DSM-IV psychotic disorder)

4 Serious risk of indirectness (participants classified as in the early initial prodromal state as opposed to a high risk mental state
Table 37: Summary of findings table for outcomes reported for integrated psychological therapy versus supportive
counselling at 104 weeks follow-up
Outcome or subgroup Study ID Number of Effect estimate (SMD Heterogeneity Quality of evidence
studies / or RR) [95% CI] (GRADE)a
participants
Transition to psychosis (RR) BECHDOLF2012 K = 1, N =125 0.32 [0.11, 0.92]* N/A Very low1,2,3
Leaving the study early for any reason (RR) BECHDOLF2012 K = 1, N = 128 0.95 [0.61, 1.49] N/A Very low1,2,3
Note. ROB = Risk of bias; RR = Relative risk; SMD = Standardised mean difference. *Favours integrated psychological therapy
aThe GRADE approach was used to grade the quality of evidence for each outcome.
1Serious risk of bias missing data).
2Optimal information size (for dichotomous outcomes, OIS = 300 events; for continuous outcomes, OIS = 400 participants) not met.
3Serious risk of indirectness (participants classified as in the early initial prodromal state as opposed to a high risk mental state and transition is defined as the
development of either attenuated/transient symptoms or a DSM-IV psychotic disorder).

Table 38: Summary of findings table for outcomes reported for integrated psychological therapy versus standard care at 52
weeks post-treatment
Outcome or subgroup Study ID Number of studies / Effect estimate (SMD Heterogeneity Quality of evidence
participants or RR) [95% CI] (GRADE)a
Completer analysis: transition to NORDONTOFT2006 K = 1, N = 67 0.24 [0.07, 0.81]* N/A Low1,2
psychosis (RR)
Positive symptoms (SMD) NORDONTOFT2006 K = 1, N = 62 -0.30 [-0.76, 0.16] N/A Low1,2
Leaving the study early for any reason NORDONTOFT2006 K = 1, N = 79 0.63 [0.22, 1.81] N/A Low1,2
(RR)
*Favours integrated psychological therapy.
1 Serious risk of bias.
Table 39: Summary of findings table outcomes reported for integrated psychological therapy versus standard care at 104
weeks post-treatment
Outcome or subgroup Study ID Number of studies / Effect estimate (SMD Heterogeneity Quality of evidence
participants or RR) [95% CI] (GRADE)a
Completer analysis: transition to NORDONTOFT2006 K = 1, N = 65 0.52 [0.26, 1.02] N/A Low1,2
psychosis (RR)
Positive symptoms (SMD) NORDONTOFT2006 K = 1, N = 57 -0.36 [-0.89, 0.16] N/A Low1,2
Negative symptoms (SMD) NORDONTOFT2006 K = 1, N = 57 -0.42 [-1.09, 0.25] N/A Low1,2
Leaving the study early for any reason NORDONTOFT2006 K = 1, N = 79 0.66 [0.25, 1.73] N/A Low1,2
(RR)
1 Serious risk of bias.

5.5.5 Clinical evidence summary for psychosocial interventions
Seven RCTs investigated the efficacy of psychological interventions in young people
at risk of developing psychosis or schizophrenia. Five trials compared CBT with
supportive counselling and the findings suggest that CBT may have a beneficial
effect on rate of transition to psychosis. However, CBT was found to be no more
effective on than supportive counselling on psychotic symptoms, depression,
psychosocial functioning and quality at life. One RCT compared integrated
psychological therapy with supportive counselling and found small effects that
integrated psychological therapy decreases transition to psychosis. Another RCT
found a similar beneficial effect of integrated psychological therapy, when compared
with standard care, on the rate of transition to psychosis at 12 months, but this
significant effect was not found at 24 months. Moreover, when dropouts in both
groups were assumed to have transitioned the significant beneficial effect of
integrated psychological therapy on transition to a DSM-IV psychotic disorder, as
opposed to an ultra-high/high risk mental state (attenuated/transient symptoms),
was lost. Integrated psychological therapy appeared no more effective than standard
treatment on positive or negative symptoms of psychosis, or dropout. Overall,
heterogeneity between samples in terms of their degree of risk for developing
psychosis, alongside the paucity and low quality of evidence, means that no robust
conclusions can be drawn.
5.6 HEALTH ECONOMIC EVIDENCE

Systematic literature review
This section adapted systematic literature review of existing economic evidence on
interventions in people at risk of psychosis from Psychosis and Schizophrenia in
Children and Young People (NCCMH, 2013 [full guideline]). The populations and
interventions in adapted literature review were deemed to be relevant by the GDG
for this guideline. Also, an update search was generated from the date of the last
search (2012 to October 2013) to identify any new existing economic evidence. The
systematic search of the economic literature undertaken for Psychosis and
Schizophrenia in Children and Young People (NCCMH, 2013 [full guideline]) identified
two eligible studies on people at risk of psychosis (Phillips et al., 2009;Valmaggia et
al., 2009). An update search for this guideline identified one more eligible study
(McCrone et al., 2013). Two studies were conducted in the UK (McCrone et al., 2013;
Valmaggia et al., 2009) and one in Australia (Phillips et al., 2009). Details on the
methods used for the systematic search of the economic literature are described in
Chapter 3. References to included studies and evidence tables for all economic
studies included in the guideline systematic literature review are presented in
Appendix 19. Completed methodology checklists of the studies are provided in
Appendix 18. Economic evidence profiles of studies considered during guideline
development (that is, studies that fully or partly met the applicability and quality
criteria) are presented in Appendix 17, accompanying the respective GRADE clinical
evidence profiles.

In the UK McCrone and colleagues (2013) developed a decision model to assess the
cost of EIS compared with standard care (SC) in young people who either have
psychotic illness, are in an ‘at risk’ mental health state or have another mental health
problem. SC was defined as care by child and adolescent mental health services
(CAMHS). In the model young people with signs of psychosis are initially referred
to CAMHS. Following referral, a decision is made to refer on to a specialist EIS team
or to continue to provide SC. If psychosis has developed, then the treatment options
were either to admit the service user to inpatient care or to provide community-
based support. If the service user was in an ‘at risk’ state, then either psychosocial
intervention, medical intervention, a combination of these or no treatment was
provided. The time horizon of the analysis was 6 months and the perspective of a
mental health services was adopted, with impacts on other health services and social
care not included. In the analysis the transition probabilities were based on various
published sources and where necessary were supplemented with authors’
assumptions. The study included medication costs, psychiatrist and psychologist
contacts, nurse/care coordinator contacts, and inpatient care. The resource use
estimates were based on various published sources; data provided by mental health
trust (that is, service monitoring records and clinical reporting system), and authors’
assumptions. The unit costs were obtained from national sources. The mean cost per
person over 6 months was £13,186 for EIS and £18,000 for SC group in 2009/10
prices. This represents a cost savings of £4,814 associated with the intervention. The
costs savings were mainly due to the reduced length of stay for those with psychosis
who were admitted. The model was robust to changes in most parameters and only
changing the probability of admission and increasing the length of stay for EIS
service users had an impact on the results; however changes in these parameters
would need to be relatively high. The analysis was judged by the GDG to be
partially applicable to this guideline review and the NICE reference case. Even
though the study was conducted in the UK, the authors have measured costs only
from the mental health service perspective, and haven’t looked at health effects. The
estimates of transition probabilities were obtained from various published sources
and where necessary were supplemented with authors’ assumptions; some of the
resources use estimates were derived from one mental health trust; and therefore
there may be issues of generalisability. Time horizon of the analysis was only 6
months which may not be sufficiently long to reflect all important differences in
costs. The authors have conducted extensive deterministic sensitivity analysis,
however due to the lack of data probabilistic sensitivity analysis was not
undertaken. Overall, this study was judged by the GDG to have potentially serious
methodological limitations.
Valmaggia and colleagues (2009) conducted a cost-effectiveness analysis of an EIS

service for people at high risk of psychosis. The study assessed Outreach and
Support in South London (OASIS), a service for people with an at risk mental state
for psychosis and schizophrenia. The service comprised information about
symptoms, practical and social support, and the offer of CBT and medication. The
early intervention was compared with care as usual, which did not include any
provision of specialised mental health interventions. The data on care as usual was

obtained from the same geographical area of south London. The decision analytic
model was developed for a period of 1 and 2 years from two perspectives (the health
sector and society).
The decision analytic model took into account the cost of the intervention and usual
care, initial GP visit, outpatient care (including contact with the community mental
health team), informal inpatient stay and formal inpatient stay. The societal
perspective also included lost productivity costs incurred during DUP. The resource
use and cost data are acquired from national published sources and the studies
reviewed.
The clinical evidence showed that EIS for people at high risk of psychosis reduced
the risk of developing psychosis, and it also reduced the DUP. These outcomes were
used as key parameters in the economic analysis. The long and short DUP were
defined as more than or less than 8 weeks of untreated psychosis.
Valmaggia and colleagues (2009) showed that probability of transition to psychosis

with an EIS is 0.20 compared with 0.35 in the case of usual care. Data from OASIS
indicate that transition takes place on average 12 months after contact with GP or
OASIS. The probability of long DUP in the intervention group (OASIS) is 0.05. This
is lower than the usual care probability of 0.80, which consequently leads to a higher
proportion of formal and informal inpatients in the usual care group.
According to the cost results, at 1 year the expected total service cost per person was
£2,596 for EIS and £724 for usual care in 2004 prices. The 1-year duration did not
capture the transition to psychosis because it was assumed to occur at 12 months
after referral. The model estimated the expected cost of intervention at £4,313 per
person and £3,285 for usual care. Including cost of lost productivity, the 2-year
model showed cost savings with expected intervention costs of £4,396 per person
and usual care of £5,357. Therefore, the perspective taken in the analysis, health
sector or societal, is important as it changes the findings of the model. Using the
reported data, the estimated incremental cost-effectiveness ratio (ICER) is £6,853 per
person of avoiding risk of psychosis in 2004 prices.
The one-way sensitivity analysis showed that the 2-year model from a societal
perspective is robust to changes in parameter values. There was no sensitivity
analysis conducted using the NHS perspective. The economic model only covered
the 2 years’ duration of the study, however psychotic disorders can be life-long. A
longer study is required to analyse whether a lower rate of transition to psychosis in
the intervention group is temporary or permanent. The lower rate of transition to
psychosis and long DUP in the intervention group could also have substantial
economic benefits accruing beyond 2 years. Another limitation of the model is that it
used data from observational studies and not from RCTs, which could affect the
robustness of the results. The settings of the service and the local cost estimates
might not be applicable to other areas. However, sensitivity analysis mitigates this
limitation and the tree model structure can be tailored to other settings and estimates

of costs and transition probabilities. The model only took into account indirect cost
of lost employment. The cost to parents and carers for unpaid care, to social care,
and to the criminal justice system might also contribute to indirect costs that are not
accounted for. Based on the above considerations the analysis was judged by the
GDG to be only partially applicable to this guideline review and the NICE reference
case; and it was also judged by the GDG to have potentially serious methodological
limitations.
Phillips and colleagues (Phillips et al., 2009) conducted a cost-minimisation study of

specific and non-specific treatment for young people at ultra-high risk of developing
first episode of psychosis in Australia. The analysis compared the costs of a specific
preventive intervention with a needs-based intervention. The specific preventive
intervention comprised a combination of risperidone and cognitively-oriented
psychotherapy in addition to ‘needs-based treatment’ (supportive counselling,
regular case management and medication) for 6 months.
The mean age of participants in both groups was 20 years. The analysis took the
perspective of the Australian healthcare sector. The costs of inpatient and outpatient
services and pharmacological interventions were calculated at the end of treatment
(at 6 months) and at 12 and 36 months’ follow-up for young people attending the
Personal Assessment and Crisis Evaluation (PACE) Clinic in Melbourne, Australia.
The costs were measured in Australian dollars in 1997 prices and the 36 months’
follow-up costs were discounted at 3%.
As the cost analysis was conducted after the completion of the trial, several
assumptions were made regarding resource use during the treatment. Resource use
was calculated via a patient questionnaire during follow-up, which could have
introduced errors. The unit costs were acquired from the budget and financial
information of the service and national published sources on mental health costs in
Australia.
The results were presented as mean costs for both groups for inpatient and
outpatient services and pharmacological interventions and total costs of the
treatment phase (6 months) and 12 and 36 month’s follow-up. The specific
preventive intervention had significantly higher cost for outpatient services of
AU$2,585 during the treatment phase compared with the needs-based intervention
of AU$1,084. However, the outpatient cost of specific preventive intervention at
36 months is AU$4,102, which is significantly lower than the needs-base intervention
cost of AU$10,423. The differences between total costs and other components of the
two intervention groups during the treatment phase and 12 and 36 months’ follow-
up were not statistically significant.
The findings of the study were not definitive; however, the analysis indicated
substantial cost savings associated with the specific preventive intervention in the
longer term. Most importantly, the study highlights that despite high outpatient
costs of the specific preventive intervention during the treatment phase and at

12 months’ follow-up, it incurred significantly lower outpatient costs than the needs-
based intervention at 36 months’ follow-up. The lower cost of the specific preventive
intervention at 36 months was not associated with the treatment outcome as there
were no differences in functioning or quality of life. The side effects of the
intervention captured in the clinical trial are not accounted for in the health
economic analysis, which could alter the findings substantially. The analysis is
valuable because it used patient-level data and compared two services of different
levels of intensity. However, the sample size of the study is small and not
representative beyond the ultra-high risk subgroup, which is a limitation. In
addition, the resource-use data were based on assumptions because the cost analysis
was conducted after the completion of the trial and the patient questionnaire at
follow-up could have led to patients erroneously recalling resource use. Based on the
above considerations the analysis was judged by the GDG to be only partially
applicable to this guideline review and the NICE reference case; and it was also
judged by the GDG to have potentially serious methodological limitations. On
reflection, the GDG concluded that the health economic analysis was unsupportable
within the context of this guideline.

Relative value placed on the outcomes considered
The GDG considered the critical outcomes to be:
• Transition to psychosis
• Time to transition to psychosis.
However, this is often a highly comorbid, help-seeking group that requires support
and treatment. Therefore, the GDG also through it pertinent to consider:
• Mental state (symptoms, depression, anxiety, mania)
• Mortality (including suicide)
• Global state
• Psychosocial functioning
• Leaving the study early for any reason
• Adverse effects (including effects on metabolism, EPS, hormonal
changes and cardiotoxicity).
We found no evidence to support the early promise of some antipsychotic drugs in

delaying or preventing transition to psychosis. In addition, antipsychotic drugs are
associated with clinically significant side effects. Although this is best described as
an absence of evidence rather than evidence of absence, this review identifies no
reason to pursue this line of enquiry. Many people at ultra-high risk will not
progress to psychosis, and we expect that any evidence indicating that the benefits
outweigh the harms in this population would have been published. Psychological

treatment might be associated with an increase in stigma and other consequences for
participants who would not develop psychosis without treatment.
When meta-analysed, there was no clear evidence to suggest that antipsychotic

medication can prevent transition. Moreover, adverse effects, specifically weight
gain, were clearly evident and indicate that the harms associated with antipsychotic
medication significantly outweigh the benefits.
Overall, the results for psychosocial interventions suggest that transition to

psychosis from a high-risk mental state may be preventable. These findings also
provide a baseline for developing future research strategies, and they highlight
treatments that have the most potential for reducing transition to psychosis. An
important additional consideration is that there is good evidence from data in adults
that family intervention is effective in reducing relapse rates in both first episode
psychosis and in established schizophrenia, providing strong empirical evidence
that the treatment strategies used here are effective in reducing the likelihood of
(subsequent) psychosis. Importantly, family intervention was a key component of
integrated psychological therapy.
Finally, one small RCT indicated that omega-3 fatty acids may also be effective in
preventing transition from at risk mental states to the development of psychosis
(even when sensitivity analysis is applied and dropouts are assumed to have
transitioned) and improving symptoms of psychosis, depression and psychosocial
functioning. Given the very small sample from which these results were obtained,
there is insufficient evidence with which to recommend the use of omega-3 fatty
acids.
Ultimately, the majority of individuals in these at risk samples do not convert to

psychosis and as a result there are serious concerns regarding the risk of exposure to
unnecessary interventions. The harms associated with intervening include stigma
and the fear of becoming psychotic (the reason why they have been included in the
trial or offered the treatment). However, the GDG considered that these risks were
acceptable if the treatments offered added no further important potential harms. The
GDG felt that, on balance, psychological treatments and the use of omega-3 fatty
acids were unlikely to be associated with other important potential harms. However,
the side effects of antipsychotic medication include weight gain, the potential for
type 2 diabetes, long-term cardiovascular disease and the risk of irreversible brain
changes resulting in effectively untreatable and permanent movement disorders
when antipsychotic drugs are used at higher dose in the long term. Given the
seriousness of these effects, that only a small proportion of individuals will go on to
develop psychosis and that the evidence suggested that antipsychotics were unlikely
to produce any benefit, antipsychotic treatment will result in unacceptable harm.
Consequently, there is a strong basis for not prescribing antipsychotic medication or
researching its use further in this population.
On the other hand, the GDG noted that because these people are treatment seeking,
often distressed and have comorbidities, they should have access to help for their
distress (CBT) and treatments recommended in NICE guidance for any comorbid
conditions such as anxiety, depression, emerging personality disorder or substance
misuse, or whatever other problem presents. Although the numbers of episodes of
psychosis prevented affect a small percentage of people at high risk of psychosis,
many others in these trials are likely to benefit from CBT for the treatment of these
other, non-psychotic psychological problems.
There were two UK-based economic studies that assessed the economic impact of
EIS for people at high risk or with signs of psychosis; however the GDG judged both
studies to have potentially serious methodological limitations. The time frame of the
analyses was very limited, however psychotic disorders can be lifelong. Also, both
studies used data from either observational studies, other published sources and
authors’ assumptions and not from RCTs. The findings of the Australian study were
not definite either. Even though it indicated potential cost savings the sample size of
the study was small and not representative beyond the ultra high-risk subgroup.
Moreover, some of resource use estimates were based on assumptions and patient
questionnaire at follow-up. As a result, the analysis was judged by the GDG to have
potentially serious methodological limitations and on reflection the GDG concluded
that the analysis was unsupportable within the context of this guideline.
Consequently, based on existing economic evidence the GDG could not draw
definite conclusions pertaining to the cost effectiveness of EIS for people at high risk
of psychosis.
For all interventions, the quality of the evidence ranged from very low to moderate.
The evidence for pharmacological interventions was of particular poor quality and
was rated as very low across all critical outcomes. A primary reason for
downgrading the quality of the evidence was risk of bias across the trials. Almost all
of the trials included in the review were rated as high risk of bias due to various
limitations within them making them difficult to interpret. Such limitations included
small sample sizes, lack of outcome assessor blinding and likely publication bias; the
latter being especially likely for antipsychotics. Furthermore, there is some
suggestion that among this high risk group, the number of transitions increases over
3 years and then settles. Therefore, trials require longer periods of follow-up. Other
reasons for downgrading the quality of evidence across interventions concerned
limited information size, indirectness or risk of reporting bias. There were also some
concerns in the definition of ‘transition to psychosis’ which varied across included
studies.
Recent studies have examined the feasibility of detecting and treating individuals
with at risk mental states, prior to the development of psychosis and schizophrenia.
Criteria are now available to identify and recognise help-seeking individuals who

are at high risk of imminently developing schizophrenia and related psychoses,
using standardised semi-structured interviews. These criteria require further
refinement in order to better predict the course of these ‘at risk’ behaviours and
symptoms, as well as recognition of those who will and those who will not go on to
develop psychosis. In addition, in order to obtain precise estimates of rates of
transition to psychosis in this population, further work is needed that looks at the
influence of sampling strategies in this population.
The GDG considered it important that people experiencing transient psychotic

symptoms or other experiences suggestive of possible psychosis were referred
urgently to a specialist mental health service where a multidisciplinary assessment
should be carried out (see recommendations 5.8.1.1 and5.8.2.1). In addition, the GDG
decided to recommend individual CBT with or without family intervention for
people at risk of developing psychosis delivered with the aim of lowering the risk of
transition to psychosis and reducing current distress (see recommendation 5.8.4.1). It
was also deemed important to monitor individuals for up to 3 years (see
recommendation 5.8.4.1), offering follow-up appointments to those who requested
discharge from the service (see recommendation 5.8.4.2). Further studies to examine
the use of family intervention to prevent a first occurrence of psychosis in those at
high risk were considered an important direction for further research.
As no evidence was found to support the early promise that some antipsychotics
may delay or prevent transition, and because antipsychotics are associated with
significant side effects, the GDG decided there was no reason to pursue this line of
enquiry, particularly since many people at ultra-high risk will not progress to
psychosis and schizophrenia (see recommendation 5.8.3.2).
5.8 RECOMMENDATIONS
5.8.1 Referral from primary care
5.8.1.1 If a person is distressed, has a decline in social functioning and has:
• transient or attenuated psychotic symptoms or
• other experiences suggestive of possible psychosis or
• a first-degree relative with psychosis or schizophrenia
refer them for assessment without delay to a specialist mental health service or
an early intervention in psychosis service because they may be at increased risk
of developing psychosis. [new 2014]
5.8.2 Specialist assessment

5.8.2.1 A consultant psychiatrist or a trained specialist with experience in at-risk
mental states should carry out the assessment. [new 2014]

5.8.3 Treatment options to prevent psychosis
5.8.3.1 If a person is considered to be at increased risk of developing psychosis (as
described in recommendation 5.8.1.1):
• offer individual cognitive behavioural therapy (CBT) with or
without family intervention (delivered as described in
recommendations 9.4.10.3 and 9.7.10.3) and
• offer interventions recommended in NICE guidance for people
with any of the anxiety disorders, depression, emerging
personality disorder or substance misuse. [new 2014]
5.8.3.2 Do not offer antipsychotic medication:
• for people considered to be at increased risk of developing psychosis (as
described in recommendation 5.8.1.1) or
• with the aim of decreasing the risk of or preventing psychosis [new 2014]
5.8.4 Monitor and follow-up

5.8.4.1 If, after treatment (as described in recommendation 5.8.3.1), the person
continues to have symptoms, impaired functioning or is distressed, but a
clear diagnosis of psychosis cannot be made, monitor the person regularly
for changes in symptoms and functioning for up to 3 years using a
structured and validated assessment tool. Determine the frequency and
duration of monitoring by the:
• severity and frequency of symptoms
• level of impairment and/or distress and
• degree of family disruption or concern. [new 2014]
5.8.4.2 If a person requests discharge from the service, offer follow-up

appointments and the option to self-refer in the future. Ask the person’s GP
to continue monitoring changes in their mental state. [new 2014]

6 ACCESS AND ENGAGEMENT
This chapter has been updated for the 2014 guideline. The review of early
intervention has been updated and is now included in Chapter 12, Teams and
service-level interventions. Sections of the guideline where the evidence has not be
updated since 2009 are marked by asterisks (**2009**_**2009**).
6.1 INTRODUCTION
**2009**Although there is great emphasis on clinical practice and service
organisation to deliver effective clinical interventions, it is well known that there are
significant social and ethnic inequalities regarding access to and benefit from such
effective clinical interventions. Schizophrenia is likely to impact negatively on
finances, employment and relationships, especially if the illness begins when the
person is very young, which is a vulnerable time and when the adverse social impact
of an illness can be most devastating. More attention is now rightly focused on
ensuring early access to effective interventions for psychosis, to reduce periods of
untreated psychosis, and also to ensure prompt and precise diagnosis, and quicker
recovery to minimise social deficits, following the onset of illness.
There is substantial evidence that patterns of inequality regarding access to and

benefit from treatment show some ethnic groups are disadvantaged and might
benefit from prompt and precise diagnosis and intervention. Furthermore, some
people from specific ethnic groups may fear services, or respond to stigma, or find
that services do not understand their personal, religious, spiritual, social and cultural
needs or their cultural identity. These needs are important for them to sustain and
maintain a healthy identity.
6.2 ACCESS AND ENGAGEMENT TO SERVICE-LEVEL

INTERVENTIONS
6.2.1 Introduction
Background and approach
Schizophrenia is known to be a devastating illness with significant social and
psychological deficits, and it is crucial that service users receive treatments and
services that are collectively sanctioned as appropriate approaches in the context of
dominant ethical, clinical and legal frameworks of practice and service organisation.
These frameworks and standards of care are informed by the evolving evidence
base and expert opinion. African–Caribbean people in the UK have been shown to
have a higher incidence of schizophrenia, while the treatment practices and service
organisation for recovery have not been especially tailored to meet their needs
(Kirkbride et al., 2006). South Asian people may also have a higher incidence of
schizophrenia, but there is less compelling evidence (Kirkbride et al., 2006).
Migrants, people living in cities, and those at the poorer and less advantaged end of
society are also at risk (Cantor-Graae & Selten, 2005). Asylum seekers and refugees
may face additional risks of poor mental health, but their experience, to date, has not
been directly linked to a higher incidence of schizophrenia, although it is related to
complex social and health needs among those developing schizophrenia (Royal
College of Psychiatrists, 2007). More generally, culture is known to influence the
content and, some would argue, the form and intensity of presentation of symptoms;
it also determines what is considered to be an illness and who people seek out for
remedy. Cultural practices and customs may well create contexts in which distress is
generated; for example, where conformity to gender, age, and cultural roles is
challenged.
Paradigms for quality improvement

The dominant paradigms for improved standards of care (including service
organisation, effective interventions, and integrated care pathways and patterns of
treatment received by ethnic groups and migrants) are the cultural psychiatry and
equalities paradigms.
The cultural psychiatry paradigm tries to understand the cultural origins of

symptoms, as well as: (a) how these symptoms are coloured when expressed across
cultural boundaries; (b) which treatments are sanctioned; and (c) whether treatments
themselves, ostensibly evidence-based, are really culturally constructed solutions
that work best for people sharing the same cultural norms and expectations of what
constitutes illness and treatment. This endeavour is largely clinically motivated and
responds to frontline evidence of a lack of appropriate knowledge and skills to
benefit all people equally using existing guidelines and treatment approaches. It also
draws upon sociology and anthropology as key disciplines.
The equalities paradigm is heavily underpinned by two national policies: Inside

Outside (National Institute for Mental Health in England, 2003) and Delivering Race
Equality (Bhui et al., 2004; Department of Health, 2003; Department of Health, 2005).
These policies promote race equality through institutional and national programmes
of actions with leadership from health authorities, mental health trusts and locally
organised groups of stakeholders. These actions have not been specific to
schizophrenia, but have certainly been motivated by the perceived crisis in the care
and treatment of African–Caribbean people with schizophrenia, to which providers
have not previously responded in a consistent and visibly effective manner. To date,
results from the Care Quality Commission’s patient census (‘Count Me In’) indicate
that policies and programmes in this area have not yet had the desired effects
(Healthcare Commission, 2008). Perceived, individual and institutional prejudice
and racism are also tackled within a broader equalities framework that addresses
multiple forms of social exclusion and stigma (McKenzie & Bhui, 2007).
Cultural competence
Encompassed in the above two paradigms is the notion of cultural competence. A
recent systematic review (Bhui et al., 2007) suggested that staff cultural competence
training may produce benefits in terms of cultural sensitivity, staff knowledge and
staff satisfaction. However, despite these promising findings, clinicians should be

aware of the problems and controversies surrounding the definition or current
understandings of cultural competence. Kleinman and Benson (2006) propose that a
cultural formulation, based upon a small scale ethnographic study of the individual
or on the DSM-IV cultural formulation, should be written for each patient. This
cultural formulation can then be used to help determine and inform appropriate
clinical interventions at the individual patient level. On the other hand, others, such
as Papadopoulous and colleagues (2004), have suggested a more model-based
approach, in which cultural competence is seen as part of a four stage conceptual
map, wherein competence is informed by and informs three other processes, namely
cultural sensitivity, cultural knowledge and cultural awareness. Whichever
approach is taken, it is clear from the literature that cultural competence is now
recognised as a core requirement for mental health professionals. Yet despite this
increased awareness of its importance, little evaluative work has been done to assess
the effects of cultural competence (at both an individual and organisational level) on
a range of service user, carer and healthcare professional outcomes.
The 2009 guideline: how did the Guideline Development Group take
account of race, ethnicity and culture?
For the 2009 guideline, the GDG did not attempt to examine all evidence relevant to
race, culture and ethnicity, but instead focused on three main approaches. First, the
two topic groups examining psychological/psychosocial interventions and
pharmacological interventions reviewed evidence of benefits for ethnic groups.
Second, where there was little evidence for specific effects for ethnic groups,
included studies (for the recommended interventions) were reviewed to assess the
ethnic diversity of the samples. This was done to establish whether the findings may
be of relevance to ethnic groups as well as the majority population. Third, a specific
topic group examining clinical questions related to access and engagement was
formed with input from special advisers. In particular, the group requested that the
literature search should cover specialist ethnic mental health services, that studies of
service-level interventions should be examined to assess the ethnic diversity of the
samples and that preliminary subgroup analyses of existing datasets should be
conducted to inform research recommendations (see Section 6.2.11).
Limitations
The focus on race, culture and ethnicity in this 2009 guideline is welcomed and
ground-breaking, but there is a limitation in the sense that all mental healthcare
should be similarly reviewed, with a broader focus. Regarding this 2009 guideline,
the methodologies developed have necessarily been targeted on some key issues and
are not comprehensive in their actions. The 2009 guideline has also not been able to
look at broader issues of pathways to care and effectiveness of psychological and
pharmacological interventions on the basis of new and different levels of evidence.
In part, this is because there is limited evidence. Furthermore, the 2009 guideline has
not looked at issues that were not reviewed in the 2002 guideline. Therefore the
following might be usefully accommodated in further reviews: matching the racial
identity of the professional with the service user, ethnic matching (which is broader
than matching racial identity and also encompasses cultural similarities), the impact

of social exclusion and racism across generations, and the impact on young people of
parents who have been socially excluded, subjected to prejudice and have a mental
illness. All of these might seem imperative to service users from black and minority
ethnic groups, but were not within the scope of the 2009 guideline. It is vital that
future guideline updates attend to these broader issues, perhaps additionally with a
guideline for these issues across disease areas.
On evidence and ethnicity

There are general concerns that current evidence relating to ethnicity has not come
from adequate samples of ethnic groups (or any socially excluded group). There are
also concerns regarding the hierarchy of evidence. First, in the absence of high-
quality evidence, expert opinion and the dominant paradigms of treatment are given
preference over other forms of evidence (for example, qualitative evidence); second,
clinical trials are given preference over other study designs. Thus, existing
institutionalised practices are sustained. Research studies propose that there are
pharmacokinetic and pharmacodynamic differences in drug handling across
migrant, national and ethnic groups, but our scientific understanding of these at an
ethnic-group level does not permit generalised statements to be made about a group
that can then be applied to the individual from that group. Psychological therapies
may privilege psychologised forms of mental distress, perhaps excluding those
experiencing social manifestations of distress that is not so easily recognised as
having a mental component. However, this 2009 guideline could not fully address
these issues.
Assuming that service users from black and minority ethnic groups can benefit from
the same interventions delivered in the same way, the next question is whether black
and minority ethnic groups have equal access to these effective interventions and
whether they remain in contact with services. The access and engagement topic
group focused on this broad question of engagement and retained contact with
existing innovative services that aim to be flexible and should be culturally
appropriate, namely assertive community treatment (assertive outreach teams),
crisis resolution and home treatment teams, and case management. For this work,
existing reviews of these services were reanalysed for data on ethnic groups with
loss to follow-up and contact with services as the primary outcome. The next part
reviews the literature for evidence that ethnic-specific or culturally-adapted services
were effective or more effective at preventing loss to follow-up, dropout and
sustained contact over time. The interventions reviewed are defined below.
Definitions
Assertive community treatment (assertive outreach teams)

The bipolar disorder guideline (NCCMH, 2006 [full guideline]) review of assertive
community treatment (ACT) updated the review undertaken for the 2002
schizophrenia guideline, which was based on the review by Marshall and Lockwood
(2002). This latter review identified the key elements of ACT as:

• a multidisciplinary team-based approach to care (usually involving a
psychiatrist with dedicated sessions)
• care is exclusively provided for a defined group of people (those with serious
mental illness)
• team members share responsibility for clients so that several members may
work with the same client and members do not have individual caseloads
(unlike case management)
• ACT teams attempt to provide all the psychiatric and social care for each
client rather than referring on to other agencies
• care is provided at home or in the work place, as far as this is possible
• treatment and care is offered assertively to uncooperative or reluctant service
users (‘assertive outreach’)
• medication concordance is emphasised by ACT teams.
The bipolar disorder guideline (NCCMH, 2006 [full guideline]) adopted the
definition of ACT used by Marshall and Lockwood (2002) which followed a
pragmatic approach based upon the description given in the trial report. For a study
to be accepted as ACT, Marshall and Lockwood (2002) required that the trial report
had to describe the experimental intervention as ‘Assertive Community Treatment,
Assertive Case Management or PACT; or as being based on the Madison, Treatment
in Community Living, Assertive Community Treatment or Stein and Test models.’
ACT and similar models of care are forms of long-term interventions for those with
severe and enduring mental illnesses. Thus, the review did not consider the use of
ACT as an alternative to acute hospital admission. The review also excluded studies
of ‘home-based care’, as these were regarded as forms of crisis intervention, and are
reviewed with crisis resolution and home treatment teams.
Crisis resolution and home treatment teams

The GDG for the bipolar disorder guideline (NCCMH, 2006 [full guideline]) adopted
the inclusion criteria developed by the Cochrane Review (Joy et al., 2002) for studies
of crisis resolution and home treatment teams (CRHTTs) in the management of
people with schizophrenia. Crisis intervention for people with serious mental health
problems was selected by the bipolar disorder GDG for review and further analysis.
Crisis intervention and the comparator treatment were defined as follows:

• Crisis resolution: any type of crisis-orientated treatment of an acute
psychiatric episode by staff with a specific remit to deal with such situations,
in and beyond ‘office hours’.
• Standard care: the normal care given to those experiencing acute psychiatric
episodes in the area concerned. This involved hospital-based treatment for all
studies included.
The focus of the review was to examine the effects of CRHTT models for anyone
with serious mental illness experiencing an acute episode when compared with the
‘standard care’ they would normally receive.

Case management
Given the variation in models of case management evaluated in the literature, the
bipolar disorder GDG adopted the definition used in a Cochrane review (Marshall et
al., 2000) where an intervention was considered to be ‘case management’ if it was
described as such in the trial report. In the original review no distinction, for
eligibility purposes, was made between ‘brokerage’, ‘intensive’, ‘clinical’ or
‘strengths’ models. For the purposes of the bipolar disorder guideline (NCCMH,
2006 [full guideline]) review, intensive case management (ICM) was defined as a
caseload of less than or equal to 15. The UK terms ‘care management’ and ‘care
programme approach’ were also treated as synonyms for case management.
However, the review excluded studies of two types of intervention often loosely
classed as ‘case management’, including ACT and ‘home-based care’.
Specialist ethnic mental health services (culturally specific or culturally skilled)

Specialist ethnic mental health services aim, by definition, to offer a culturally
appropriate service and effective interventions to either a specific racial, ethnic,
cultural or religious group or to deliver an effective service to diverse ethnic groups
(Bhui et al., 2000; Bhui & Sashidharan, 2003). Models of specialist services have not
been mapped recently but include cultural consultation service styles, and others
outlined by Bhui and colleagues (2000).
6.2.2 Clinical review protocol

The review protocol, including the primary clinical question, information about the
databases searched and the eligibility criteria can be found in Table 40. For the 2009
guideline, all studies were examined for information about ethnicity of the sample
and numbers losing contact with services by ethnic group. The access and
engagement topic group and special advisers developing the guideline proposed
that a sample of which at least 20% of subjects were from black and minority ethnic
groups could be considered ‘ethnically diverse’. It was assumed that a decrease in
the number of participants leaving the study early for any reason indicated that the
service was more engaging.
Table 40: Clinical review protocol for the review of services
Primary clinical questions For all people from black and minority ethnic groups (particularly,
African–Caribbean people) with psychosis, do services, such as ACT,
CRHTTs and case management improve the number of people
remaining in contact with services?
For all people from black and minority ethnic groups with psychosis,
do specialist ethnic mental health services (culturally specific or
culturally skilled) improve the number of people remaining in contact
with services?
Electronic databases MEDLINE, EMBASE, PsycINFO, CINAHL
Date searched Database inception to 6 April 2008

Other resources searched Bipolar disorder guideline (NCCMH, 2006) and reference lists of
included studies
Study design Any
Patient population People with psychosis from a black and minority ethnic group in the
UK
Interventions 1. ACT, CRHTTs and case management
2. Specialist ethnic mental health services (culturally specific or
culturally skilled)
Outcomes Number of people remaining in contact with services (measured by
the number of people lost to follow-up or loss of engagement with
services)
However, the GDG acknowledges that people may leave a study early for reasons
other than a lack of engagement with the service.
6.2.3 Studies considered for review

Assertive community treatment (assertive outreach teams)
The bipolar disorder guideline (NCCMH, 2006 [full guideline]) included 23 RCTs of
ACT: 13 versus standard care (N = 2,244), four versus hospital-based rehabilitation
(N = 286) and six versus case management (N = 890). Studies included had to
conform to the definition of ACT given above, and the inclusion criteria used by
Marshall and Lockwood (2002) were widened to include populations with serious
mental illness.
Of the 23 trials included in the bipolar disorder guideline (NCCMH, 2006 [full
guideline]), nine included adequate information about ethnicity of the sample,
although none reported outcome data by ethnic group. Therefore, the GDG
conducted a sensitivity analysis of seven studies that had an ethnically diverse
sample (see Table 41 for further information).

The bipolar disorder guideline (NCCMH, 2006 [full guideline]) included seven RCTs
of a CRHTT versus inpatient care (N = 1,207). Of these, three included an ethnically
diverse sample, and one (MUIJEN1992) reported the number of people leaving the
study early for any reason by ethnicity (see Table 42 for further information).
Case management
The bipolar disorder guideline (NCCMH, 2006 [full guideline]) review updated the
review undertaken for the 2002 schizophrenia guideline and included 17 RCTs of
case management: 13 versus standard care (intensive and standard case
management [SCM]), two intensive versus standard case management, one
enhanced case management versus standard case management and one case
management versus brokerage case management. One trial (BRUCE2004) was
excluded from the present review as 100% of participants had a diagnosis of
depression. Of the 16 remaining RCTs, six included an ethnically diverse sample,
and three of these studies (FRANKLIN1987; MUIJEN1994; BURNS1999) reported the

number of people leaving the study early for any reason by ethnicity (see Table 42
for further information).
Specialist ethnic mental health services

For the 2009 guideline, papers were included in the review if they reported
comparisons of UK-based specialist mental-health service interventions and/or
initiatives. An inclusive definition of ‘specialist ethnic service’ was used to include
those services that were either culturally adapted or tailored to the needs of
individual patients, including any religious or ethnic needs. To measure improved
access and engagement, the numbers of people from different black and minority
ethnic groups remaining in contact with services (as measured by loss to follow-up
and loss of engagement) was the primary outcome. All study designs were
considered and papers were included even if a formal evaluation of the service had
not been intended.
Papers were excluded from the review if: (a) they only reported descriptions of
current service use by different black and minority ethnic groups, (b) did not report
any comparison between services, and (c) were non-UK based or did not report loss
to follow-up/ loss of engagement within different black and minority ethnic groups.
The reference lists of included papers and any relevant reviews were further checked
for additional papers. The review was restricted to English language papers only.
The search identified 2,284 titles and abstracts, of which 19 were collected for further
consideration. All 19 papers were excluded because of lack of comparator, failure to
report loss to follow-up and/or loss of engagement by ethnicity or were non- UK
interventions.

6.2.4 Assertive community treatment or crisis resolution and home
treatment teams versus control
Table 41: Study information and evidence summary table for trials of ACT or
CRHTTs
ACT versus ACT versus ACT versus CRHTTs versus standard

standard care hospital-based case care
rehabilitation management
k (total N) 5 RCTs (N = 684) 1 RCT (N = 59) 1 RCT (N = 3 RCTs (N = 492)

Study ID AUDINI1994 CHANDLER1997 BUSH1990 FENTON1998
BOND1998 MUIJEN1992
BOND1990 PASAMANICK
LEHMAN1997 1964
MORSE1992
Diagnosis 30–61% 61% 86% 49–100%

schizophrenia schizophrenia schizophrenia schizophrenia
Ethnicity AUDINI1994: 26% 40% African– 50% black FENTON1998: 14%

African–Caribbean American (ACT), black (CRHTTs),
BOND1998: 34% 55.2% African– 28% black (control)
black, American (control) MUIJEN1992: 25% African–
2% Latino Caribbean (CRHTTs), 21%
BOND1990: 30% African–Caribbean (control)
black PASAMANICK
LEHMAN1997: 1964: 32.9%
61% African– non-white
American (ACT),
84% African–
American
(control)
MORSE1992: 52.5%
non-white (mostly
African–American)
Outcomes
Leaving the RR 0.63 (0.48, 0.82), RR 1.55 (0.28, RR not RR 0.73 (0.43,
study early k = 5, N = 684, I 2 = 8.62), k = 1, N = 59 estimable 1.25), k = 3,
for any 0% (nobody left N = 492, I2 = 57%
reason the study
Excluding studies early) Excluding
targeting homeless PASAMANICK
people: RR 0.62 1964: RR 0.66 (0.50, 0.88), k =
(0.44, 0.89), k = 3, N 2, N = 374, I2 = 0%
= 416, I2 = 0%

Leaving the African–
study early Caribbean: RR 1.12 (0.51,
for any 2.45), k = 1, N = 43
reason by Other non-white: RR 0.70
black and (0.21,
minority 2.34), k = 1, N = 26
group
6.2.5 Case management versus control
Table 42: Study information and evidence summary table for trials of case
management
Standard case Intensive case ICM versus SCM

management management (ICM)
(SCM) versus versus standard
Total number 1 RCT (N = 413) 4 RCTs (N = 362) 1 RCT (N = 708)
of studies
(number of
participants)
Study ID FRANKLIN1987 FORD1995 BURNS1999(UK700) 10
HOLLOWAY1998
MUIJEN1994
SOLOMON1994
Diagnosis 56% 66–83% 87% schizophrenia or
schizophrenia schizophrenia schizoaffective disorder
10Subgroup by ethnicity data obtained from authors.

Ethnicity 25% black, 2% FORD1995: 23% 29% African–Caribbean,
Hispanic (SCM), black 20% other black and minority
24% black, and minority ethnic ethnic groups (ICM) 26%
6% Hispanic groups (ICM), 37% African– Caribbean, 20% other
(control) black and minority black and minority ethnic groups
ethnic groups (SCM)
(control)
HOLLOWAY1998:
51% non-white
(ICM), 57%
non-white (control)
MUIJEN1994: 29%
African–Caribbean,
2% Asian (ICM), 17%
African–Caribbean,
5% Asian (control)
SOLOMON1994: 83%
black, 3% Hispanic
Outcomes
Leaving the RR 0.95 (0.74, RR 0.76 (0.53, 1.09), k RR 0.56 (0.38, 0.82),
study early for 1.23), = 4, N = 362, I² = 3.9% k = 1, N = 708
any reason k = 1, N = 413,
Leaving the - Black: RR 0.74 (0.48, White: RR 0.73 (0.38,

study early for 1.23), 1.40), k = 1, N = 267
any reason by k = 2, N = 121 African–Caribbean: RR
black and 1.00 (0.53, 1.87), k = 1, N = 270
minority
ethnic group
Lost contact - - RR 1.71 (1.09, 2.69),

with case k = 1, N = 708
manager
Refused - - RR 1.44 (0.55, 3.73),
contact with k = 1, N = 708
case manager
6.2.6 Secondary subgroup analyses

Given the paucity of evidence available to answer questions about the use of, and
engagement with, services by people from black and minority ethnic groups, the
GDG examined data from two service-level intervention studies conducted in the
UK (Johnson et al., 2005; Killaspy et al., 2006). Patient-level data were made available
to the GDG during the development of the guideline for the purposes of conducting
secondary post hoc analyses to examine loss of contact and engagement with the
service by ethnicity of the participants. These analyses were exploratory in nature
and were intended to be purely hypothesis generating as opposed to generating
evidence to underpin recommendations. Both studies were non-blind RCTs (see
Table 43 for further details).

In both trials, participants categorised as black African, black Caribbean or black
other were included in the black and minority ethnic subgroup. Additionally, in the
North Islington Crisis study (Johnson et al., 2005) participants categorised as ‘mixed
race’ were included in the subgroup analysis. As far as possible, the same
procedures used in the primary papers were applied to the secondary analysis
conducted for this 2009 guideline. For example, where a primary paper excluded
missing data, the same procedure was subsequently applied to the present analysis.
In addition to looking at engagement with services as measured by numbers losing
contact, other measures of access and engagement (including contact with forensic
services and engagement rating scales) were included in the present analysis. For
continuous measures, because of the high potential for skewed data, Mann Whitney-
U tests were applied to test for differences in the median values. For dichotomous
outcomes, Chi-squared tests were applied where appropriate to test for differences
with relative risks calculated for variables such as relapse and rehospitalisation.
Although the main findings are summarised below, more detailed evidence tables
for each subgroup comparison can be found in Appendix 23b.
REACT (Killaspy et al., 2006)

The findings can be summarised as follows:
• In the whole sample, there was no difference in the proportion consenting to
treatment in the group of participants allocated to ACT versus standard care.
This finding was replicated in the subgroup of black and minority ethnic
participants.
• In the whole sample, ACT was associated with reduced loss to follow-up at
both
• 9 and 18 months. These findings were not demonstrated in the subgroup of
black and minority ethnic participants.
• In the whole sample, ACT improved service user engagement, but this
finding did not hold for black and minority ethnic subgroup.
• In both the whole sample and the black and minority ethnic subgroup, ACT
• increased the number of contacts with mental health professionals at both 9
and
• 18 months.
• ACT had no effect on any measure of detention or hospitalisation (including
involuntary admissions) in both the whole sample and the black and minority
ethnic subgroup.

Table 43: Details of studies included in the secondary subgroup analyses
Study Objective Design/ Setting Participants Groups Main outcome

measures
REACT To compare outcomes of Non-blind RCT/two 251 men and women Intervention = treatment Primary outcome was
(Killaspy et care from ACT with care inner London boroughs under the care of adult from ACT team (127 inpatient bed use 18
al., by CMHTs for people secondary mental health participants) months after
2006) with serious mental services with recent high randomisation. Secondary
illnesses use of inpatient care and Comparator = outcomes included
difficulties engaging with continuation of care from symptoms, social
community services CMHT (124 participants) function, client
satisfaction, and
engagement with
services.
North To evaluate the Non-blind RCT/ London 260 residents of the inner Intervention = acute care Primary outcome was
Islington effectiveness of a crisis borough of Islington London borough of including a 24- hour crisis hospital admission and
Crisis RCT resolution team Islington who were resolution team number of inpatient bed
experiencing crises severe (experimental group) use. Secondary
(Johnson enough for hospital outcomes included
et al., admission to be Comparator = standard symptoms and client
2005) considered care from inpatient satisfaction.
services and CMHTs
(control group)
Note. ACT = assertive community treatment; CMHT = community mental health team

North Islington Crisis team RCT (Johnson et al., 2005)
The findings can be summarised as follows:
• The crisis team intervention significantly reduced hospitalisation rates and

number of inpatient bed days for both the whole sample and the black and
minority ethnic subgroup.
• The crisis team intervention had no impact on treatment compliance or
numbers lost to follow-up, for both the whole sample and the black and
minority ethnic subgroup.
• The number of professional contacts, including contacts with GPs increased at
8 weeks and 6 months, and although the effect was not significant in the black
and minority ethnic subgroup, the point estimate suggests this is because of a
small sample size and resulting lack of statistical power, rather than the
absence of an effect.
• For both the sample as a whole and the black and minority ethnic subgroup,
the crisis team intervention did not impact upon any measure of involuntary
detention or status under the Mental Health Act.
6.2.7 Other sources of evidence

The review of ethnically-specific or adapted services yielded no UK-based studies
that investigated loss to follow-up. However, some of the studies, although falling
outside the guideline’s inclusion criteria, offer important lessons for clinical practice
and research. Bhugra and colleagues (2004) demonstrated that black people in
contact with mental health services via contact with either primary care or non-
primary care services were equally as dissatisfied as a white group gaining access to
services from outside primary care. The most satisfied group were identified as
white people accessing mental health service following contact and referral from
primary care. Mohan and colleagues (2006) showed, in a non-randomised study, that
subsequent to the introduction of intensive case management, black patients were
more likely to have greater contact with psychiatrists and nurses, while white
patients more often had greater social care contact. Black patients were less likely to
require hospital admission. Khan and colleagues (2003) showed in a small
qualitative study that South Asian people receiving care from a home treatment
team valued the intervention because of the cultural appropriateness in terms of
language, religious needs, dietary needs and stigma, while hospitals were preferred
for investigations (for example, blood tests).
A systematic review of interventions that improve pathways into care for people
from black and minority ethnic groups was recently completed (Moffat et al., 2009;
Sass et al., 2009). This was commissioned by the Department of Health through the
Delivering Race Equality programme (established in 2005). The systematic grey
literature search yielded 1,309 documents, of which eight fully met inclusion criteria.
The main findings of the review indicated that:

‘The key components of effective pathway interventions include specialist
services for ethnic minority groups, collaboration between sectors,
facilitating referral routes between services, outreach and facilitating access
into care, and supporting access to rehabilitation and moving out of care.
Services that support collaboration, referral between services, and improve
access seem effective, but warrant further evaluation. Innovative services
must ensure that their evaluation frameworks meet minimum quality
standards if the knowledge gained from the service is to be generalised, and
if it is to inform policy’ (Moffat et al., 2009).
The review of mainstream published literature identified 2,216 titles and abstracts
with six studies meeting the review’s inclusion criteria. In only one study was the
initiative UK based, and included patients with depression as opposed to psychosis.
The main findings of the review indicated that
‘There was evidence that interventions led to three types of pathways

change; accelerated transit through care pathways, removal of adverse
pathways, and the addition of a beneficial pathway. Ethnic matching
promoted desired pathways in many groups but not African Americans,
managed care improved equity, a pre- treatment service improved access to
detoxification and an education leaflet increased recovery’ (Sass et al., 2009).
In addition to these findings, the review concluded that further research is needed to
facilitate evidence-based guidance for the development of services.

Although there were no RCTs assessing the effectiveness of ACT for specific ethnic
groups, five RCTs including an ethnically diverse sample indicated that when
compared with standard care ACT interventions were effective in reducing loss to
follow-up. When compared with standard care alone, CRHTTs were also effective at
reducing loss to follow-up. Only one RCT (MUIJEN1992) included in the review
permitted stratification of these effects by ethnic group. The positive findings from
this RCT regarding reduced loss to follow-up held most strongly for Irish people,
but was not convincing for African–Caribbean subgroups. However, it must be
noted that because of the limited sample size no firm conclusions can be drawn from
this one RCT alone. The review of case management included more RCTs permitting
stratification of outcomes by ethnicity. Despite this, there was no consistent evidence
for the effectiveness of either intensive or standard case management when
compared with standard care and other service configurations.
Although the search of specialist ethnic mental health services undertaken for the
2009 guideline did not yield any eligible studies, recent reviews (Moffat et al., 2009;
Sass et al., 2009) both grey and mainstream literature provided some interesting
examples of how cultural adaptations can lead to improved outcomes. However it
must be noted that even within these reviews, there was paucity of information, with
the majority of included studies being non-UK based, thus limiting the

generalisability to specific black and minority ethnic populations within the
UK.**2009**
6.2.9 Linking evidence to recommendations

The systematic review for the 2009 guideline did not provide any robust evidence to
warrant changing the service recommendations in the 2002 guideline for people with
schizophrenia from black and minority ethnic groups. However, the GDG for the
2009 guideline and the special advisers recognised that there were a number of
problems specifically faced by people from different black and minority ethnic
groups, including:
• **2009**People from black and minority ethnic groups with schizophrenia are
more likely than other groups to be disadvantaged or have impaired access to
and/or engagement with mental health services.
• People from black and minority ethnic groups may not benefit as much as
they could from existing services and interventions, with the aforementioned
problems in access and engagement further undermining any potential
benefits.
• For all people with a first episode of psychosis or severe mental distress
(including those from black and minority ethnic groups), fears about the
safety of the intervention may not be appropriately addressed by the clinician.
• Conflict may arise when divergent explanatory models of illness and
treatment expectations are apparent.
• Clinicians delivering psychological and pharmacological interventions may
lack an understanding of the patient’s cultural background.
• The lack of supportive and positive relationships may impact on the future
engagement with services.
• Comprehensive written information may not be available in the appropriate
language.
• Participants from black and minority ethnic groups may face additional
language barriers with a lack of adequate interpretation services being
available. Where such services are available, clinicians may lack the training
to work proficiently with such services.
• Lack of knowledge about the quality of access for specific black and minority
ethnic groups and inflexible approaches to service delivery may hamper
continued engagement with treatment.
• There is often a lack of collaborative work between mental health service
providers and local voluntary and charitable sectors that may have expertise
in the provision of the best cultural or specific services.
• Race, culture, ethnicity or religious background may challenge the clarity
with which assessments and decisions regarding the Mental Health Act are
undertaken, especially where clinicians do not seek appropriate advice
and/or consultation.**2009**
Therefore, based on informal consensus, the GDG for the 2009 guideline made
recommendations that address, in at least an initial way, the problems raised above.

Additionally, where possible, specific problems faced by black and minority ethnic
groups have been addressed in other parts of the guideline (for example, see Section
9.7.6).
The recommendations from the 2009 guideline remain but because of the change in
population addressed by the 2014 guideline the recommendations have been
changed to reflect this to say ‘people with psychosis or schizophrenia’
It was further acknowledged by the GDG for the 2009 guideline that all of the
recommendations in this section should be viewed as a foundation step in a longer
process including the provision of good quality research and development. In
particular, the GDG highlighted that the following points specifically need
addressing through this process of research:
• **2009**RCTs of psychological and pharmacological interventions and service

organisation have not been adequately powered to investigate effects in
specific ethnic groups including African–Caribbean people with
schizophrenia.
• There are no well-designed studies of specialist mental health services
providing care to diverse communities or to specific communities.
• The effect of the cultural competence of mental health professionals on service
user experience and recovery has not been adequately investigated in UK
mental health settings.
• English language teaching may be an alternative to providing interpreters to
reduce costs and to encourage integration. This has not been tested for
feasibility or outcomes.
• The early diagnosis and assessment of psychosis and comorbid disorders
across ethnic, racial and cultural groups needs to be systematically assessed,
with research projects including adequate samples from different cultural and
ethnic backgrounds. **2009**
Following publication of Service User Experience in Adult Mental Health (NICE, 2011),
one recommendation about communication and provision of information, which
was covered by that guideline, was removed.

6.2.10 Recommendations
6.2.10.1 Healthcare professionals inexperienced in working with people with
psychosis or schizophrenia from diverse ethnic and cultural backgrounds
should seek advice and supervision from healthcare professionals who are
experienced in working transculturally. [2009]
6.2.10.2 Healthcare professionals working with people with psychosis or
schizophrenia should ensure they are competent in:
• assessment skills for people from diverse ethnic and cultural backgrounds
• using explanatory models of illness for people from diverse ethnic and
cultural backgrounds
• explaining the causes of psychosis or schizophrenia and treatment options
• addressing cultural and ethnic differences in treatment expectations and
adherence
• addressing cultural and ethnic differences in beliefs regarding biological,
social and family influences on the causes of abnormal mental states
• negotiating skills for working with families of people with psychosis or
schizophrenia
• conflict management and conflict resolution. [2009]
6.2.10.3 Mental health services should work with local voluntary black, Asian and
minority ethnic groups to jointly ensure that culturally appropriate
psychological and psychosocial treatment, consistent with this guideline and
delivered by competent practitioners, is provided to people from diverse
ethnic and cultural backgrounds. [2009]
6.2.11 Research recommendations

6.2.11.1 For people with schizophrenia, RCTs of psychological and psychosocial
interventions should be adequately powered to assess clinical and cost
effectiveness in specific ethnic groups (or alternatively in ethnically diverse
samples). [2009]
6.2.11.2 An adequately powered RCT should be conducted to investigate the clinical
and cost effectiveness of CBT that has been culturally adapted for African–
Caribbean people with schizophrenia where they are refusing or intolerant
of medication.[2009]
6.2.11.3 Studies of ethnically specific and specialist services and new service designs
should be appropriately powered to assess effectiveness. Studies should
include sufficient numbers of specific ethnic groups and be evaluated using
an agreed high quality evaluation framework (Moffat et al., 2009).[2009]

6.2.11.4 For people with schizophrenia from black and minority ethnic groups living
in the UK, does staff training in cultural competence at an individual level
and at an organisational level (delivered as a learning and training process
embedded in routine clinical care and service provision) improve the service
user’s experience of care and chance of recovery, and reduce staff burnout?
[2009]
6.2.11.5 An adequately powered proof of principle study should be conducted to
investigate the feasibility of comparing language skills development for
those with English as a second language against using interpreters. [2009]
6.2.11.6 A study should be conducted to investigate engagement and loss to follow-
up, prospective outcomes and care pathways, and the factors that hinder
engagement. For example, ethnic, religious, language or racial identity
matching may be important. This is not the same as ethnic matching, but
matching on ability to work with diverse identities.[2009]
6.2.11.7 A study should be conducted to investigate the use of pre-identification
services, including assessment, diagnosis and early engagement, across
racial and ethnic groups.[2009]

7 INTERVENTIONS TO PROMOTE
PHYSICAL HEALTH IN ADULTS
7.1 INTRODUCTION
This chapter is new for the 2014 guideline and aims to review the evidence for
interventions that promote physical health in adults with psychosis and
schizophrenia. For the purpose of this guideline, this chapter is divided into two
sections. The first (Section 7.2) is concerned with behavioural interventions to
promote physical activity and healthy eating, while the second (Section 7.3) assesses
the efficacy of interventions for reducing and stopping smoking.
7.2 BEHAVIOURAL INTERVENTIONS TO PROMOTE

PHYSICAL ACTIVITY AND HEALTHY EATING
7.2.1 Introduction
For people with psychosis and schizophrenia, a combination of poor diet and
nutrition, weight gain and lack of physical activity are important contributors to
high rates of physical comorbidities such as type 2 diabetes and reduced life
expectancy particularly from cardiovascular disease. Moreover weight gain and
obesity further contribute to stigma and discrimination and may explain unplanned
discontinuation of antipsychotic medication leading to relapse.
Since the 2009 guideline (NICE, 2009d) a greater emphasis on prevention is indicated
by increasing evidence that adverse effects associated with an increased risk of long-
term health problems are prevalent with the use of antipsychotics (Newcomer et al.,
2013). Additionally, cardiometabolic risks appear within weeks of commencing
antipsychotics, particularly weight gain and hypertriglyceridaemia and later glucose
dysregulation and hypercholesterolemia (Foley & Morley, 2011). The importance of
prevention is further emphasised by evidence that over a third of people with
established schizophrenia taking antipsychotics can be identified biochemically to be
at high risk of diabetes (Manu et al., 2012). Indeed this group was specifically
highlighted by NICE in its guidance on preventing type 2 diabetes, in which lifestyle
interventions are recommended followed by metformin if lifestyle approaches are
not successful (NICE, 2012c).
Developing recommendations about lifestyle interventions is hampered by a paucity

of evidence, particularly large or longer-term studies or in people with first episode
psychosis. The limited research has mainly been directed towards weight reduction
rather than physical activity programmes, although in practice these approaches
may overlap. A recent systematic review evaluated non-pharmacological
interventions to reduce weight for people using antipsychotic medication
(Caemmerer et al., 2012). The review observed a mean weight reduction of 3.12 kg

over a period of 8 to 24 weeks. Clinically significant reductions in waist
circumference and improvements in cardiovascular risk factors were also shown.
The benefits were seen irrespective of the duration of treatment, whether the
intervention was delivered to an individual or in a group setting, and whether the
intervention was based on CBT or a nutritional intervention. In addition, outpatient
programmes appeared to be more effective than inpatient programmes. Weight
reduction should not be the only concern since poor nutrition may directly
contribute to physical ill health for this population. Again, however, there is a
paucity of evidence about interventions to address these issues.
7.2.2 Clinical review protocol (behavioural interventions to promote

physical activity and healthy eating)
review protocols can be found in Appendix 6; further information about the search
The review strategy was to evaluate the clinical effectiveness of the interventions
using meta-analysis. However, in the absence of adequate data, the available
evidence was synthesised using narrative methods.
Table 44: Clinical review protocol summary for the review of behavioural
interventions to promote physical activity and healthy eating
Review For adults with psychosis and schizophrenia, what are the benefits and/or
question(s) potential harms of behavioural interventions to promote physical activity (all
forms, with or without healthy eating)?
For adults with psychosis and schizophrenia, what are the benefits and/or
potential harms of behavioural interventions to promote healthy eating?
Objectives To evaluate the clinical effectiveness of interventions to improve the health of
people with psychosis and schizophrenia.
Population Adults (18+) with schizophrenia (including schizophrenia-related disorders such
as schizoaffective disorder and delusional disorder) or psychosis.
Intervention(s) • Behavioural interventions to promote physical activity (with or without
healthy eating)
• Behavioural interventions to promote healthy eating
Comparison Any alternative management strategy
Critical outcomes • Physical health
• BMI/ weight
• Levels of physical activity
• Service use
• Primary care engagement (for example, GP visits)
• Quality of life
• User satisfaction (validated measures only)
Electronic CORE: CDSR, CENTRAL, DARE, Embase, HTA, MEDLINE, MEDLINE In-
database process
Topic specific: CINAHL, PsycINFO

Date searched RCT: database inception to June 2013
SR: 1995 to June 2013
Study design RCT
• End of treatment
• Up to 6 months’ follow-up (short term)
• 7-12 months’ follow-up (medium term)
• 12 months’ follow-up (long term)
Where more than one follow-up point within the same period was available, the
latest one was reported.
Sub-analysis
Where data were available, sub-analyses were conducted of studies with ≥75% of
the sample described as having a primary diagnosis of schizophrenia/
schizoaffective disorder or psychosis.
Where data were available, sub-analyses were conducted for UK/Europe

studies.

Twenty four RCTs (N = 1972) met the eligibility criteria for this review (see the sub-
sections below). All studies were published in peer-reviewed journals between 1978
and 2013. Further information about both included and excluded studies can be
found in Appendix 15a.
The trials identified evaluated the effectiveness of behavioural interventions to

promote physical activity in combination with healthy eating and interventions to
promote physical activity alone. No studies with the singular aim of promoting
healthy eating were identified. Table 45 provides an overview of the trials included
in each category.
Behavioural interventions to promote physical activity and healthy

eating
Of the eligible trials, 15 RCTS (N = 1,337) evaluated a combined behavioural physical
activity and healthy eating intervention compared with an alternative management
strategy: ALVAREZ2006 (Alvarez-Jiménez et al., 2006), ATTUX2013 (Attux et al.,
2013), BRAR2005 (Brar et al., 2005), BROWN2011 (Brown et al., 2011), DAUMIT2013
(Daumit et al., 2013), EVANS2005 (Evans et al., 2005), KWON2006 (Kwon et al.,
2006), LITTRELL2003 (Littrell et al., 2003), MAURI2008 (Mauri et al., 2008),
MCKIBBIN2006 (McKibbin et al., 2006), SCOCCO2006 (Scocco et al., 2006),
SKRINAR2005 (Skrinar et al., 2005), WU2007 (Wu et al., 2007), WU2008 (Wu et al.,
2008) and USHER2012 (Usher et al., 2013).

All 15 trials followed a psychoeducation/information-based approach and provided
information and support for how to increase levels of physical activity and healthy
eating. Four of the included trials (DAUMIT2013, SKRINAR2005, WU2007, WU2008)
additionally included prescribed physical activity as a part of the intervention.
Participants in the intervention arm of one trial (WU2008) were prescribed
metformin (N=64) 12. Of the 15 trials, 13 included a large proportion (≥75%) of
participants with a primary diagnosis of psychosis or schizophrenia. None of the
included trials were based in the UK. Table 45 provides an overview of the included
trials.
Behavioural interventions to promote physical activity

Of the eight eligible trials (N = 635), seven (N = 455) evaluated a behavioural
physical activity intervention compared with an alternative management strategy:
ACIL2008 (Acil et al., 2008), BEEBE2010 (Beebe, 2010), CHAO2010 (Chao, 2010),
COLE1997 (Cole, 1997), PAJONK2010 (Pajonk et al., 2010), SCHEEWE2013 (Scheewe
et al., 2013) and VARAMBALLY2012 (Varambally et al., 2012); two trials (N = 180)
evaluated one type of physical activity intervention with another programme:
DURAISWAMY2007 (Duraiswamy et al., 2007) and VARAMBALLY2012.
VARAMBALLY2012 was used in both comparisons.
Five of the seven eligible trials (ACIL2008, COLE1997, PAJONK2010,

SCHEEWE2013, VARAMBALLY2012) included prescribed physical activity as an
integral part of the intervention. A single trial (BEEBE2010) provided participants
with information about physical activity and another (CHAO2010) provided
participants with a pedometer that was used and monitored in daily life for the
prescribed period. Two trials (DURAISWAMY2007, VARAMBALLY2012) evaluated
a yoga intervention versus an aerobic training programme.
Of the eligible trials, six included a large proportion (≥75%) of participants with a
primary diagnosis of psychosis or schizophrenia. None of the included trials was
based in the UK. Table 45 provides an overview of the included trials.
12An oral diabetes medication that is used to control blood sugar levels.

Table 45: Study information table for trials included in the meta-analysis of
behavioural interventions to promote physical activity and healthy eating versus
any alternative management strategy
Physical activity and Physical activity Physical activity

healthy eating interventions versus (yoga) versus
interventions versus any any alternative physical activity
alternative management management strategy (aerobic)
strategy
Total no. of trials (k); k = 15 ; N = 1337 k = 7; N = 455 k = 2; N = 180
participants (N)
Study ID(s) ALVAREZ2006 ACIL2008 DURAISWAMY2007
ATTUX2013 BEEBE2010 VARAMBALLY20123
BRAR2005 CHAO2010
BROWN2011 COLE1997
DAUMIT2013 PAJONK2010
EVANS2005 SCHEEWE2013
KWON2006 VARAMBALLY2012
LITTRELL2003
MAURI2008
MCKIBBIN2006
SCOCCO2006
SKRINAR2005
USHER2012
WU2007
WU2008
Country Australia (k =2) Germany (k = 1) India (k = 2)
Brazil (k = 1) India (k = 1)
China (k =2) Netherlands (k = 1)
Italy (k =2) Turkey (k = 1)
South Korea (k =1) USA (k =3)
Spain (k =1)
USA (k =6)
Year of publication 1996 to 2013 1997 to 2012 2007 to 2012
Mean age of 38.35 years (26.3 to 54 36.41 years (29.7 to 46.9 31.9 years (32.6 to
participants (range) years)1 years) 32.3 years)
Mean percentage of 87.46% (10.2 to 100%)2 83.19% (21.7 to 100%) 100% (100 to 100%)
participants with
primary diagnosis of
psychosis or
schizophrenia (range)
Mean gender % 50.56% (24.6 to 68.8%) 39.84% (0% to 74.6%) 31.1% (30.3 to 30.7%)
women (range)
Length of treatment 8 to 26 weeks 2 to 26 weeks 3 to 4 weeks
Length of follow-up End of treatment only End of treatment only Up to 6 months
ATTUX2013 ACIL2008 DURAISWAMY2007
BRAR2005 CHAO2010 VARAMBALLY2012
BROWN2011 COLE1997
KWON2006 PAJONK2010
MAURI2008 SCHEEWE2013
MCKIBBIN2006
SCOCCO2006 Up to 6 months
SKRINAR2005 BEEBE2010
USHER2012 VARAMBALLY2012
WU2007

WU2008
Up to 6 months
ALVAREZ2006
DAUMIT2013
EVANS2005
LITTRELL2003
MCKIBBIN2006
Up to 12 months
ALVAREZ2006
DAUMIT2013
Intervention type Achieving Healthy Aerobic exercise Yoga- Swami
Lifestyles in Psychiatric training (k =2) Vivekananda
Rehabilitation (ACHIEVE) Exercise therapy (k = 1) Yoga Anusandhana
(k = 1) Pedometer with and Samsthana (k = 2)
Behavioural weight-loss without self-monitoring
treatment (k = 1) (k = 1)
Diabetes Awareness and Physical activity
Rehabilitation Training programme (k = 1)
(DART) (k = 1) Physical exercise:
Early behavioural adopted from the
intervention (k = 1) National Fitness Corps’
Healthy lifestyle Handbook for Middle
intervention (k =3) High and Higher
Lifestyle Wellness Secondary Schools (k = 1)
Program (k = 1) WALCS group
Nutrition education education sessions (k =
sessions (k = 1) 1)
Passport 4 Life Yoga - Swami
programme (k = 1) Vivekananda Yoga
Psychoeducation class - Anusandhana
Solutions of Wellness Samsthana (k = 1)
modules (k = 1)
Psychoeducational
intervention and referral
to a nutritionist (k = 1)
Psychoeducational
Program (PEP) for weight
control (k = 1)
Recovering Energy
Through Nutrition and
Exercise for Weight Loss
(RENEW) (k = 1)
Weight management
programme (k = 1)
Comparisons Information booklet (k = 1) No pedometer control Physical exercise:
No treatment - waitlist (k (k = 1) adopted from the
= 1) Occupational therapy (k National Fitness
Olanzapine treatment as = 1) Corps’ Handbook for
usual (k = 3) Table top football (k = Middle High and
Passive nutritional 1) Higher Secondary
education from the booklet Time-and-attention Schools (k = 2)
'Food for the Mind' (k = 1) control (k = 1)
Standard care (k =8) Treatment as usual (k =
Usual care plus 3)

information (k = 1)
Note. WALCs = Walk, Address Sensations, Learn About Exercise, Cue Exercise for schizophrenia spectrum
disorders.
1 One study (USHER2012) failed to report mean age.
2 One study (SKRINAR2005) failed to report % diagnosis.
3 VARAMBALLY2012 was composed of three arms and was used in both ‘physical activity interventions versus
any alternative management strategy’ and ‘physical activity (yoga) versus physical activity (aerobic)’
comparisons.
7.2.4 Clinical evidence for behavioural interventions to promote

physical activity and healthy eating
presented in Table 46, Table 47 and Table 48. The full evidence profiles and
associated forest plots can be found in Appendix 17 and Appendix 16, respectively.
Behavioural interventions to promote physical activity and healthy

eating
Low quality evidence from up to 14 trials (N = 1,111) showed that a behavioural
physical activity and healthy eating intervention had a significant effect on reducing
body weight at the end of treatment and at short-term follow-up. There was no
difference between the intervention and control groups at short-term follow-up for
weight reduction. There was inconsistent evidence for changes in activity level.
Moderate to low quality evidence from up to six trials with 353 participants showed
that behavioural interventions to promote physical activity and healthy eating had a
small but significant positive effect on quality of life and participant satisfaction at
the end of treatment. No data evaluating this at follow-up were identified.
None of the trials evaluated provided data for the crucial outcome of primary care
engagement.
Sub-analysis (psychosis and schizophrenia only)

For the critical outcomes of body weight/BMI, the sub-analysis findings did not
differ from the main analysis. Unlike the main analysis, there is no evidence of an
increase in quality of life in favour of the active intervention. No other critical
outcome data were available. See Appendix 16 for the related forest plots.

Table 46: Summary of findings table for trials of physical activity and healthy
eating interventions compared with any alternative management strategy
Outcomes Illustrative comparative risks* (95% No. of participants Quality of the evidence
CI) (studies) (GRADE)
Corresponding risk
Physical activity and healthy eating
Body mass Mean body mass (weight end of 1,111 ⊕⊕⊝⊝
(weight) - end of treatment) in the intervention groups (14 studies) Low1,2
treatment was 2.8 lower (3.6 to 1.99 lower)
Body mass Mean body mass-(weight up to 6 449 ⊕⊕⊝⊝
(weight) - up to months’ follow-up) in the (5 studies) Low1,3
6 months’ intervention groups was 2.33 lower
follow-up (3.31 to 1.34 lower)
Body mass Mean body mass (weight > 12 247 ⊕⊕⊕⊝
(weight) - > 12 months’ follow-up) in the (1 study) Moderate1
months’ follow- intervention groups was 3.20 lower
up (5.17 to 1.23 lower)
Quality of life - Mean quality of life (end of 353 ⊕⊕⊝⊝
end of treatment treatment) in the intervention groups (6 studies) Low1,3
was 0.24 standard deviations higher
(0.01 to 0.47 higher)
Satisfaction - Mean satisfaction (end of treatment) 71 ⊕⊕⊕⊝
end of treatment in the intervention groups was 0.75 (1 study) Moderate4
standard deviations higher (0.26 to
1.23 higher)
Physical health Mean physical health (CGI activity 34 ⊕⊕⊝⊝
(exercise) - end level end of treatment) in the (1 study) Low3,4
of treatment - intervention groups was 1.04
Clinical Global standard deviations higher (0.28 to
Impression 1.81 higher)
(CGI): activity
Level
Physical health Mean physical health (total minutes 57 ⊕⊕⊝⊝
(exercise) - end of activity end of treatment) in the (1 study) low3,4
accelerometry standard deviations higher (0.03 to
(total minutes of 1.09 higher)
activity)
Physical health Mean physical health (IPAQ-short 126 ⊕⊕⊕⊕
(exercise) - end score end of treatment) in the (1 study) High
International standard deviations higher (0.34
Physical lower to 0.36 higher)
Activity
Questionnaire-
short version
(IPAQ-short)
Physical health Mean physical health (total minutes 52 ⊕⊕⊝⊝
(exercise) - up to of activity up to 6 months’ follow-up) (1 study) Low3
6 months’ in the intervention groups was 0.22
follow-up – standard deviations higher (0.33
accelerometry lower to 0.76 higher)
(total minutes of

activity)
1 Most studies included are at moderate risk of bias.
2 Evidence of serious heterogeneity of study effect size.
3 CI crosses clinical decision threshold.
4 Crucial limitation for one criterion or some limitations for multiple criteria sufficient to lower confidence in the
estimate of effect.
Behavioural interventions to promote physical activity
Physical activity versus any alternative management strategy

There was no conclusive evidence favouring physical activity over control for
reducing weight, quality of life or increasing levels of physical activity as measured
by a researcher. However, one trial (N = 53), using a subjective self-report, presented
moderate quality evidence of an increase in physical activity for the intervention
group at the end of the intervention, but this was not maintained at short-term
follow-up.
None of the included trials provided data for the critical outcomes of primary care
engagement and user satisfaction.

For the critical outcome of physical activity levels, the sub-analysis findings did not
differ from the main analysis. No other critical outcome data were available. See
Appendix 16 for the related forest plots.
Physical activity (yoga) versus physical activity (aerobic)

One trial (N = 41) presented high quality evidence that yoga when compared with
aerobic physical activity improved quality of life at short-term follow-up. No other
critical outcomes were reported for this review.

For the critical outcome of quality of life, the sub-analysis findings did not differ
substantially from the main analysis. No other critical outcome data were available.
See Appendix 16 for the related forest plots.

Table 47: Summary of findings table for physical activity interventions compared
with any alternative management strategy
Outcomes Illustrative comparative risks* (95% CI) No. of Quality of

Corresponding risk participants the
Physical activity (studies) evidence
(GRADE)
Physical health Mean physical health (weight end of treatment) in 105 ⊕⊝⊝⊝
(weight/BMI) - end of the intervention groups was 0.20 higher (0.20 (2 study) Very low1,2,3
treatment lower to 0.59 higher)
Quality of life - end of Mean quality of life (end of treatment) in the 83 ⊕⊝⊝⊝
treatment intervention groups was (2 studies) Very
0.62 standard deviations higher (0.41 lower to 1.66 low1,2,4,5
higher)
Physical activity Mean physical activity (minutes walked end of 97 ⊕⊕⊝⊝
(minutes walked) - end of treatment) in the intervention groups was (1 study) Low2,6
treatment 0.24 standard deviations higher (0.16 lower to 0.64
higher)
Physical activity (IPAQ- Mean physical activity (IPAQ-short score) in the 53 ⊕⊕⊕⊝
short telephone format) intervention groups was (1 study) Moderate6
0.32 standard deviations higher (0.27 lower to 0.91
higher)
Physical activity Mean physical activity (minutes walked up to 6 97 ⊕⊕⊝⊝
(minutes walked) - up to months’ follow-up) in the intervention groups (1 study) Low2,6
6 months’ follow-up was 0.34 standard deviations higher (0.06 lower to
0.74 higher
1 Concern as to the applicability of intervention and population.
2 CI crosses the clinical decision threshold (SMD of 0.2 or -0.2; RR of 0.75 or 1.75).
3 Suspicion of publication bias.
4 Most information is from studies at moderate risk of bias.
5 Evidence of very serious heterogeneity of study effect size.
estimate of effect.
Table 48: Summary of findings table for yoga compared with aerobic exercise

Physical activity (yoga) evidence
(GRADE)
Quality of life - up to 6 Mean quality of life (up to 6 months’ 41 ⊕⊕⊕⊕
months’ follow-up follow-up) in the intervention groups was (1 study) High
0.34 standard deviations higher (0.06 lower
to 0.74 higher)
footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the
relative effect of the intervention (and its 95% CI).

Overall the evidence suggests that behavioural interventions to promote physical
activity and healthy eating are effective in reducing body weight/BMI and this effect
can be maintained in the short term. As no longer-term data were available, the
effects greater than 6 months are not known. There is no consistent evidence (across
outcome rater types) of a beneficial effect on the levels of physical activity. In
addition, there is evidence that an intervention that combines a behavioural
approach to promoting both physical activity and healthy eating can improve
quality of life when measured at the end of treatment. However, the longer-term
benefits are not known. In sub-analysis including trials with a majority sample of
participants with a primary diagnosis of psychosis or schizophrenia, the findings did
not differ from the main analysis.
Interventions that aimed to promote physical activity alone were not found to be any
more effective than control in reducing weight/BMI, with inconclusive evidence
with regards to increased levels of physical activity. Additionally there was no
evidence of an increase in quality of life at the end of treatment. Limited evidence
suggests that a yoga intervention is more effective than aerobic physical activity in
improving quality of life in the short term. These findings did not differ for the
psychosis and schizophrenia subgroup.
7.2.6 Health economics evidence

No studies assessing the cost effectiveness of behavioural interventions to promote
physical health in people with psychosis and schizophrenia were identified by the
systematic search of the economic literature undertaken for this guideline. One study
currently in press (Winterbourne et al., (2013a) was identified following information
provided by the GDG. Details on the methods used for the systematic search of the
economic literature are described in Chapter 3. References to included studies and
evidence tables for all economic studies included in the guideline systematic
literature review are provided in Appendix 19. Completed methodology checklists
of the studies are provided in Appendix 18. Economic evidence profiles of studies
considered during guideline development (that is, studies that fully or partly met the
applicability and quality criteria) are presented in Appendix 17, accompanying the
respective GRADE clinical evidence profiles.
Winterbourne and colleagues (2013a) performed a cost-utility analysis comparing a

3-month intervention involving psychoeducation, nutritional and/or exercise
counselling with standard care. Standard care involved basic advice on weight and
exercise, on the risk of developing a cardiovascular event and/or type 2 diabetes
mellitus and life expectancy. A hypothetical cohort of 1000, 30-year old male service
users with first episode psychosis was modelled in yearly cycles over their lifetime.
In the first cycle, following the weight-gain prevention intervention, these
individuals could either remain in a health state where baseline weight gain is
unchanged or gain 7% of their initial bodyweight. In addition, in every cycle, the
service users can transition to a health state where they have diabetes and/or a

major cardiovascular event. The analysis was performed from the perspective of the
UK NHS and adopted a lifetime perspective. Only direct healthcare costs were
included in the analysis and the primary outcome measure was the QALY. The
expected mean lifetime costs per person were £6,893 and £6,293 for the intervention
and standard care groups, respectively. According to the model the mean lifetime
QALYs were 14.0 and 13.4 for the intervention and standard care groups,
respectively. The cost per QALY associated with the intervention was £960, which is
far below NICE’s lower cost-effectiveness threshold value of £20,000. Moreover, the
cost- effectiveness acceptability analysis showed that at a willingness to pay of
£20,000 per QALY, the probability of the intervention being cost effective was 0.95.
Deterministic sensitivity analysis found the cost per QALY to be sensitive to the
intervention effect, intervention costs and utility values. Using alternative 12-month
follow-up data, where transition probability from baseline to weight gain health
state increased from 0.26 to 0.78 and the cost of the intervention increased from £856
to £1,288, resulted in the intervention being dominated by standard care. A range of
subgroup analyses were performed (that is, changing gender, smoking status,
baseline BMI and diagnosis). However, in all of the sub-analyses the cost per QALY
was in the range of £705-1,034. Overall the analysis was judged to be partially
applicable to this guideline review and the NICE reference case. Even though it
excluded costs relevant to the PSS perspective the authors reported that these were
expected to account only for a small proportion of the total NHS and social care costs
(<10%) for people with psychosis and schizophrenia and so are unlikely to affect the
results. Also, it is not clear whether the definition of standard care is applicable to
the current practice in the NHS as it was adapted from the studies included in the
meta-analyses of the intervention effect. Moreover, diabetes and CVD risk estimates
were based on risk algorithms for the general population. Research in people with
mental health problems indicate that they are at higher risk than the general
population of certain physical health problems including obesity (Hert et al., 2011),
which in turn leads to higher risk of cardiovascular disease and diabetes. The
authors have partially allowed for higher risk in this population by assuming that
people in the cohort were heavy smokers. The utility values were taken from UK
population but the EQ-5D ratings were from a mix of UK, German and US patient
samples. The resource utilisation was based on RCT data and authors’ assumptions,
which may limit the generalisability of the findings. As a result, this analysis was
judged by the GDG to have potentially serious methodological limitations.

The GDG agreed that the main aims of a physical health and/or healthy eating
intervention should be to improve health, reduce weight and improve quality of life
(Sattelmair et al., 2011; Tuomilehto et al., 2011). The GDG also considered the
importance of engaging the service user in the intervention. Therefore, the GDG
decided to focus on the following, which were considered to be critical:
• physical health
• BMI/ weight

• levels of physical activity
• service use
• primary care engagement (for example, GP visits)
• quality of life
• user satisfaction (validated measures only).

A wealth of research in the general population supports the importance of being
physically active and having a healthy, balanced diet. For adults with psychosis and
schizophrenia, interventions that aim to both increase physical activity and improve
healthy eating are effective in reducing weight. Although data assessing benefits in
the short and long term were sparse, the evidence suggested benefits are sustained.
Furthermore, both improved quality of life and satisfaction with the intervention
were observed. The GDG considered this evidence of clinical benefit to be of
particular importance in a population with greatly increased risk of mortality.

The health economic evidence on interventions to promote physical health in adults
with psychosis and schizophrenia was limited to one UK study. Despite the study’s
limitations (for instance, lack of robust long-term clinical evidence and the model not
considering the potential savings to the NHS as a consequence of reducing other
obesity-related illnesses), the results provide evidence that non-pharmacological
interventions that include psychoeducation, nutritional and/or exercise counselling,
can be successful in preventing weight gain in the short term in people with
psychosis and schizophrenia. The positive economic finding supports the GDG’s
view that these interventions are not only of important clinical benefit but also are
likely to be cost effective within the NICE decision-making context.

The evidence ranged from very low to high across both groups of interventions. For
the combined physical health and healthy eating intervention, evidence was of better
quality and rated from low to moderate across critical outcomes. Reasons for
downgrading included risk of bias, inconsistency (although the direction of effect
was consistent across studies) and, for some outcomes, imprecision.
The review of behavioural interventions that promote healthy eating (without a
physical activity component) did not identify any studies meeting the review
protocol. The evidence suggests that a behavioural intervention to increase physical
activity and healthy eating is effective in reducing weight and improving quality of
life in adults with psychosis and schizophrenia. The GDG considered the possibility
of cross-referring to existing guidance in this area for the general population.
However, people with psychosis and schizophrenia are at a high risk of morbidity
and mortality because of physical complications such as diabetes, obesity,
cardiovascular disease and other related illness. Therefore, the GDG decided it was

important to generate recommendations specifically for this population and felt the
available evidence assisted in informing these recommendations. They did, however,
see the benefit of making specific reference to NICE guidance on obesity and
prevention of diabetes and cardiovascular disease.
Evidence suggests that long periods of mild physical activity, for example walking,
is more effective than shorter periods of moderate to vigorous exercise in improving
insulin action and plasma lipids for people who are sedentary. The GDG
purposefully decided to use the terms ‘physical activity ’and ‘healthy eating’ (rather
than the potentially stigmatising words ‘exercise’ and ‘diet’) in order to take this
evidence into consideration and promote a long-term lifestyle change rather than a
short-term ‘fix’ to reduce weight (Duvivier et al., 2013).
The GDG went beyond the evidence of clinical benefit to consider other important
issues that can determine the physical health of an adult with psychosis or
schizophrenia. These issues relate to when physical health problems should be
assessed, how they should be monitored and who should be responsible for both
physical and mental health. The GDG considered and discussed the important role
of primary care in monitoring physical health (especially current diabetes and
cardiovascular disease) and that this should be made explicit in the care plan. The
GDG believed that these issues were of equal importance to the service user’s health
as the interventions themselves.
Finally, two recommendations from the 2009 guideline, which were developed by
GDG consensus and originally included in the chapter on service–level interventions
(which has been updated for the 2014 guideline), have also been included here.
7.2.8.1 People with psychosis or schizophrenia, especially those taking
antipsychotics, should be offered a combined healthy eating and physical
activity programme by their mental healthcare provider. [new 2014]
7.2.8.2 If a person has rapid or excessive weight gain, abnormal lipid levels or
problems with blood glucose management, offer interventions in line with
relevant NICE guidance (see Obesity [NICE clinical guideline 43], Lipid
modification [NICE clinical guideline 67] and Preventing type 2 diabetes
[NICE public health guidance 38]. [new 2014]
7.2.8.3 Routinely monitor weight, and cardiovascular and metabolic indicators of
morbidity in people with psychosis and schizophrenia. These should be
audited in the annual team report. [new 2014]
7.2.8.4 Trusts should ensure compliance with quality standards on the monitoring
and treatment of cardiovascular and metabolic disease in people with
psychosis or schizophrenia through board-level performance indicators.
[new 2014]

7.2.8.5 GPs and other primary healthcare professionals should monitor the physical
health of people with psychosis or schizophrenia when responsibility for
monitoring is transferred from secondary care, and then at least annually.
The health check should be comprehensive, focusing on physical health
problems that are common in people with psychosis and schizophrenia.
Include all the checks recommended in 10.11.1.3 and refer to relevant NICE
guidance on monitoring for cardiovascular disease, diabetes, obesity and
respiratory disease. A copy of the results should be sent to the care
coordinator and psychiatrist, and put in the secondary care notes. [new
2014]
7.2.8.6 Identify people with psychosis or schizophrenia who have high blood
pressure, have abnormal lipid levels, are obese or at risk of obesity, have
diabetes or are at risk of diabetes (as indicated by abnormal blood glucose
levels), or are physically inactive, at the earliest opportunity following
relevant NICE guidance (see Lipid modification [NICE clinical guideline 67],
Preventing type 2 diabetes [NICE public health guidance 38], Obesity [NICE
clinical guideline 43], Hypertension [NICE clinical guideline 127],
Prevention of cardiovascular disease [NICE public health guidance 25] and
Physical activity [NICE public health guidance 44]). [new 2014]
7.2.8.7 Treat people with psychosis or schizophrenia who have diabetes and/or
cardiovascular disease in primary care according to the appropriate NICE
guidance (for example, see Lipid modification [NICE clinical guideline 67],
Type 1 diabetes [NICE clinical guideline 15], Type 2 diabetes [NICE clinical
guideline 66], Type 2 diabetes – newer agents [NICE clinical guideline 87]).
[2009]
7.2.8.8 Healthcare professionals in secondary care should ensure, as part of the care
programme approach, that people with psychosis or schizophrenia receive
physical healthcare from primary care as described in recommendations
12.2.5.7, 7.2.8.5–7.2.8.7. [2009]

7.2.9.1 What are the short- and long-term benefits to physical health of guided
medication discontinuation and/or reduction in first episode psychosis and
can this be achieved without major risks? [2009]

7.3 INTERVENTIONS FOR SMOKING CESSATION AND
REDUCTION
7.3.1 Introduction
A UK community cohort study (Brown et al., 2010) of people with schizophrenia
found that 73% smoked, that smoking-related disease accounted for 70% of the
excess natural mortality in the cohort, and that the risk of mortality was doubled for
those who smoked. These high rates contrast with around only 22% of the general
population who currently smoke (The NHS Information Centre & Lifestyles
Statistics, 2011).
Interventions for smoking cessation in the general population range from basic
advice to more intensive approaches involving pharmacotherapy coupled with
either individual or group psychological support; the three main pharmacotherapies
are nicotine replacement therapy (NRT), bupropion (antidepressant) and varenicline
(a nicotinic receptor partial agonist) (Campion et al., 2008). Banham and Gilbody
(Banham & Gilbody, 2010) reviewed eight RCTs of pharmacological and/or
psychological interventions for smoking cessation for people with severe mental
illness (schizophrenia and bipolar disorder). In their review most cessation
interventions showed moderate benefit, some reaching statistical significance. The
authors concluded that treating tobacco dependence was effective and those
treatments that work in the general population also work for those with severe
mental illness and appear approximately equally effective. These trials observed few
adverse events, nor were adverse effects on psychiatric symptoms noted, most
significant changes favouring the intervention groups over the control groups.
Notwithstanding these potential benefits smokers with severe mental illness are
rarely referred to smoking cessation services (Campion et al., 2008).
7.3.2 Clinical review protocol (interventions for smoking cessation

and reduction)
guideline, can be found in Table 49 (a complete list of review questions and their
related protocols can be found in Appendix 6; further information about the search

Table 49: Clinical review protocol summary for the review of interventions for
smoking cessation and reduction
Review question For adults with psychosis and schizophrenia, what are the benefits and/or
potential harms of interventions for smoking cessation and reduction?
Objectives To evaluate the clinical effectiveness of interventions to improve the health of
people with psychosis and schizophrenia
Population Adults (18+) with schizophrenia (including schizophrenia-related disorders
such as schizoaffective disorder and delusional disorder) or psychosis
Intervention(s) Included interventions
Only pharmacologcial inteventions that aim for smoking reduction or
cessation will be evaluated. These include:
• bupropion
• varenicline
• transdermal nicotine patch.
Excluded interventions
This review will not evaluate:
• interventions that report smoking outcomes but the primary aim is
not smoking reduction or cessation
• non-pharmacological interventions (because they are already
addressed in other guidelines)
• combined non-pharmacological and pharmacological interventions.
Critical outcomes • Anxiety and depression
• Physical health
• Smoking (cessation or reduction)
• Weight/BMI
• Quality of life
Electronic databases CORE: CDSR, CENTRAL, DARE, Embase, HTA, MEDLINE, MEDLINE In-
process
Date searched • RCT: database inception to June 2013
• SR: 1995 to June 2013
Study design RCT
• 6-8 weeks’ follow-up (short-term)
• Up to 6 months’ follow-up (medium-term)
Greater than 6 months’ follow-up (long-term)
Analyses were conducted for follow-up using data from the last follow-up
point reported within the time-point groupings.
Sub-analysis
Where the data were available, sub-analyses were conducted of studies with
>75% of the sample described as having a primary diagnosis of
schizophrenia/ schizoaffective disorder or psychosis.

studies.

The GDG selected an existing Cochrane review (Tsoi et al., 2013) as the basis for this
section of the guideline, with a new search conducted to update the existing review.
The existing review included 34 RCTs evaluating a variety of interventions and
comparisons. A number of these were outside the scope of this guideline, therefore,
only the comparisons relevant to this guideline are reported.
In total, 11 RCTs (N = 498) met the eligibility criteria for this review14:
+Akbarpour2010 (Akbarpour et al., 2010), +Bloch 2010 (Bloch et al., 2010), *Evins
2001 (Evins et al., 2001), *Evins 2005 (Evins et al., 2005), *Evins 2007 (Evins et al.,
2007), +Fatemi2005 (Fatemi et al., 2005), *George 2002 (George et al., 2002), *George
2008 (George et al., 2008), *Li 2009 (Li et al., 2009), *Weiner 2011 (Weiner et al., 2011),
*Weiner 2012 (Weiner et al., 2012), *Williams 2007 (Williams et al., 2007), *Williams
2012 (Williams et al., 2012a). Two trials meeting eligibility criteria were reported
only as letters to the editors or conference proceedings (+Fatemi 2005; *Williams
2007) and thus findings are described narratively. Nine studies meeting eligibility
criteria (+Akbarpour2010, +Bloch 2010, *Evins 2001, *Evins 2005, *Evins2007 ,
*George 2002, *George 2008, *Li 2009, *Weiner 2012) were published in peer-
reviewed journal. All included trials were published between 2001 and 2012. Further
information about both included and excluded studies can be found in Tsoi et al.
(2013).
Of the included trials, seven (N = 344) involved a comparison of bupropion versus

placebo with the aim of smoking cessation. Three trials (N = 103) also compared
bupropion with placebo but with the aim of smoking reduction. Two trials (N = 60)
compared varenicline with placebo with the aim of smoking cessation. One trial
compared high dose (42 mg daily) versus regular dose (21 mg daily) transdermal
nicotine patch (TNP) for smoking cessation 15. Table 50 provides an overview of the
trials included in each category.
13Changes have not been made to the study ID format used in the Cochrane review utilised in this section.
14 Studies prefixed with an asterisk (*) indicate interventions for smoking cessation and studies prefixed with a
cross (+) indicate interventions for smoking reduction.
15 This review did not evaluate two trials of TNP where treatment was for only 32 hours (Dalack GW, Meador-
Woodruff JH. Acute feasibility and safety of a smoking reduction strategy for smokers with schizophrenia.
Nicotine & Tobacco Research. 1999;1:53-7.) and 7 hours (Hartman N, Leong GB, Glynn SM, Wilkins JN, Jarvik
ME. Transdermal nicotine and smoking behavior in psychiatric patients. American Journal of Psychiatry.
1991;148:374-5. Also, patients in both trials had no desire to reduce or stop smoking.

Table 50: Study information table for trials comparing interventions for smoking
cessation and to reduce smoking with any alternative management strategy
Bupropion Bupropion versus Varenicline High dose (42 mg)

versus placebo placebo (smoking verses placebo versus regular dose
(smoking reduction) (smoking (21 mg) TNP
cessation) cessation) (smoking cessation)
Total no. of trials k =7; (N = 344) k =3; (N = 103) K=2 (N = 137) k = 1; (N = 51)
(k); participants
(N)
Study ID(s) *Evins 2001 +Akbarpour 2010 *Weiner 2011 *Williams 2007
*Evins 2005 +Bloch 2010 *Williams 2012
*Evins 2007 +Fatemi 2005
*George 2002
*George 2008
*Li 2009
*Weiner 2012
Country China (k = 1) Iran (k = 1) USA (k = 1) USA (k = 1)
USA (k = 6) Israel (k = 1) USA & Canada
USA (k = 1) (k = 1)
Year of publication 2001 to 2012 2005 to 2010 2001 to 2012 2007
Mean age of 43.46 years (38- 44.5 years (41.6- 41.1 years (not N/A3
participants 48.7 years) 47.4 years)2 reported k = 1)
(range)
Mean percentage of 100% (100 - 100%) 100% (100 - 100%) 100% (100 - 100% (100 - 100%)
participants with 100%)
primary diagnosis
of psychosis or
schizophrenia
(range)
Mean percentage of 29.62% (0 - 12.3%(0 - 24.59%)2 23% (not N/A3
women (range) 43.75%)1 reported k = 1)
Length of 4 to 12 weeks 3 to 14 weeks 12 weeks 8 weeks
treatment
Length of follow-up End of treatment End of treatment End of treatment End of treatment only
only only *Weiner 2011 *Williams 2007
*Weiner 2012 +Akbarpour 2010 *Williams 2012
+Bloch 2010
Up to 6 months +Fatemi 2005 24 weeks
*Evins 2001 *Williams 2012
*Evins 2005
*Evins 2007
*Li 2009
6- 12 months
*George 2002
*George 2008
Intervention type Bupropion (k = 7) Bupropion (k = 3) Varenicline (k = TNP 42 mg daily (k =
2) 1)
Comparisons Placebo (k = 7) Placebo (k = 3) Placebo (k = 2) TNP 21 mg daily (k =
1)
Note. TNP = transdermal nicotine patch.
1Evins 2007 did not provide data.
2 Fatemi 2005 did not provide data.
3Williams 2007 did not provide data.

7.3.4 Clinical evidence for interventions for reducing smoking
reduction or cessation
Bupropion for smoking cessation
Low to moderate quality evidence from up to seven studies (N = 340) showed that
bupropion was more effective than placebo for smoking abstinence at the end of the
intervention at up to 6 months’ follow-up.
Low to moderate quality evidence from up to four studies (N = 169) showed that
bupropion was more effective than placebo for smoking reduction (as measured by
exhaled carbon monoxide levels and cigarettes per day) at the end of treatment. No
significant difference was observed between groups at 6 months’ follow-up.
No difference between bupropion and placebo groups was reported for either
positive or negative psychosis symptoms or depressive symptoms.
Bupropion for smoking reduction

No significant difference between bupropion and placebo was observed for smoking
reduction (as measured by exhaled carbon monoxide levels) and positive or negative
psychosis symptoms at the end of the intervention.
Varenicline for smoking cessation

Low quality evidence from up to two studies (N = 137) showed that varenicline was
more effective than placebo for smoking abstinence at up to 6 months’ follow-up. No
significant difference was observed between groups at the end of the intervention.
Transdermal nicotine patch for smoking cessation

The trial evaluating this comparison was reported in a conference paper and could
be included in meta-analysis. The authors reported that there was no significant
difference between high and regular dose TNP in time to first relapse.
Summary of findings can be found in Table 51 and Table 52. The full GRADE
evidence profiles and associated forest plots can be found in Appendix 17 and
Appendix 16, respectively.

Table 51: Summary of findings table for bupropion verses placebo for smoking
cessation and reduction
Outcomes Illustrative comparative risks* (95% CI) Relative No. of Quality

Assumed Corresponding risk effect participants of the
risk (95% (studies) evidence
Control Bupropion versus placebo CI) (GRADE)
Abstinence - 6 months’ Study population RR 2.19 104 ⊕⊕⊝⊝
follow-up (primary 38 per 83 per 1000 (0.5 to (3 studies) Low1,2
outcome) - bupropion 1000 (19 to 363) 9.63)
versus placebo
36 per 79 per 1000
1000 (18 to 347)
Abstinence - 6 months’ Study population RR 3.41 110 ⊕⊕⊕⊝
follow-up (primary 36 per 124 per 1000 (0.87 to (2 studies) Moderate2
outcome) - bupropion + 1000 (32 to 484) 13.3)
TNP versus placebo + TNP
39 per 133 per 1000
1000 (34 to 519)
Abstinence - end of Study population RR 2.92 110 ⊕⊕⊝⊝
treatment (secondary 109 per 319 per 1000 (0.75 to (2 studies) Low2,3
outcome) - bupropion + 1000 (82 to 1000) 11.33)
TNP versus placebo + TNP
113 per 330 per 1000
1000 (85 to 1000)
Abstinence - end of Study population RR 3.67 230 ⊕⊕⊕⊝
treatment (secondary 52 per 191 per 1000 (1.66 to (5 studies) Moderate4
outcome) - bupropion 1000 (87 to 425) 8.14)
versus placebo
63 per 231 per 1000
1000 (105 to 513)
Reduction (expired CO N/A Mean reduction (expired CO N/A 150 ⊕⊕⊕⊝
level) - end of treatment level at the end of treatment) in (3 studies) Moderate5
(secondary outcome) - the intervention groups was
abstinence studies - studies 6.01 lower (10.2 to 1.83 lower)
using final measurements
Reduction (expired CO N/A Mean reduction (expired CO N/A 19 ⊕⊕⊝⊝
level) - the end of treatment level at the end of treatment) in (1 study) Low5
using change from baseline
Reduction (expired CO N/A Mean reduction (expired CO N/A 104 ⊕⊝⊝⊝
level) - 6 months’ follow-up level at 6 months’ follow-up) in (2 studies) Very
(secondary outcome) - the intervention groups was low2,6
abstinence studies - studies 2.08 lower (17.76 lower to 13.59
using final measurements higher)
Reduction (expired CO N/A Mean reduction (expired CO N/A 19 ⊕⊕⊝⊝
level) - 6 months’ follow-up level at 6 months’ follow-up) in (1 study) Low5
using change from baseline
Reduction (change in N/A Mean reduction (change in N/A 184 ⊕⊝⊝⊝
number of CPD from number of CPD from baseline at (3 studies) Very
baseline) - end of treatment the end of treatment) in the low1,3,5
(secondary outcome) - intervention groups was 10.77
abstinence studies lower (16.52 to 5.01 lower)

Reduction (change in N/A Mean reduction (change in N/A 104 ⊕⊕⊝⊝
number of CPD from number of CPD from baseline at (2 studies) Low2,5
baseline) - 6 months’ follow- 6 months’ follow-up) in the
up (secondary outcome) - intervention groups was
abstinence studies 0.4 higher (5.72 lower to 6.53
higher)
Reduction (change in N/A Mean reduction (change in N/A 93 ⊕⊕⊝⊝
number of CPD from number of CPD from baseline at (2 studies) Low1,2
baseline) - end of treatment the end of treatment) in the
(secondary outcome) - intervention groups was
reduction studies 2.61 lower (7.99 lower to 2.77
higher)
Note. CI = confidence interval; RR = risk ratio; CO = carbon monoxide; CPD = cigarettes per day.
5 Optimal information size not met.
Table 52: Summary of findings table for varenicline versus placebo for smoking
cessation
Outcomes Illustrative comparative risks* (95% Relative No. of Quality of the

CI) effect participants evidence
Control Bupropion versus
placebo
Abstinence – 6 months’ Study population RR 5.06 128 ⊕⊕⊝⊝
follow-up (primary 23 per 1000 118 per 1000 (0.67 to (1 study) low1,2
outcome) (16 to 889) 38.24)
23 per 1000 116 per 1000
(15 to 880)
Abstinence - end of Study population RR 4.74 137 ⊕⊕⊝⊝
treatment (secondary 42 per 1000 197 per 1000 (1.34 to (2 study) low1,2
outcome) (56 to 696) 16.71)
23 per 1000 109 per 1000
(31 to 384)
Note. CI = confidence interval; RR = risk ratio; CO = carbon monoxide; CPD = cigarettes per day.
estimate of effect.
2 CI crosses the clinical decision threshold.

This review suggests that bupropion is an effective intervention for smoking
cessation in adults with psychosis and schizophrenia immediately post-intervention
and at longer-term follow-up (up to 6 months). However, the evidence is of poor
quality and inconclusive because of the low number of studies, especially for longer-
term follow-up, resulting in wide confidence intervals. This review did not find any
adverse effects on mental state, suggesting that bupropion is well tolerated in adults
with psychosis and schizophrenia. There is no consistent evidence for the
effectiveness of bupropion for smoking reduction. There is some evidence that it is
effective in reducing smoking at the end of the intervention for both those who
attempted abstinence but did not succeed, and those who initially aimed to reduce
smoking. However, this effect is not maintained at longer-term follow-up. Limited
evidence suggests that varenicline is an effective intervention for smoking cessation
in adults with psychosis and schizophrenia at longer-term follow-up (up to 6
months) but this effect was not found immediately post-intervention. Although there
was no significant difference between the intervention and control group in
psychiatric symptoms, there were reports of suicidal ideation and behaviours from
two participants in the varenicline group. Limited evidence suggests that there is no
difference between a high and regular dose transdermal nicotine patch for smoking
cessation.

No studies assessing the cost effectiveness of interventions for reducing smoking in
people with psychosis and schizophrenia were identified by the systematic search of
the economic literature undertaken for this guideline. One study currently in press
(Winterbourne et al., 2013b) was identified following information provided by the
GDG. Details on the methods used for the systematic search of the economic
literature are described in Chapter 3. References to included studies and evidence
tables for all economic studies included in the guideline systematic literature review
are presented in Appendix 19. Completed methodology checklists of the studies are
provided in Appendix 18. Economic evidence profiles of studies considered during
guideline development (that is, studies that fully or partly met the applicability and
quality criteria) are presented in Appendix 17, accompanying the respective GRADE
clinical evidence profiles.
Winterbourne and colleagues (2013b) conducted a cost-utility analysis comparing

bupropion in combination with CBT and NRT with standard care (defined as CBT
and NRT only) in service users with psychosis and schizophrenia. In a Markov
model, a hypothetical cohort of 1000, 27-year old male smokers, was modelled in 6-
monthly cycles over their lifetime. In each cycle, smokers could quit, thus becoming
former smokers, or they could remain smokers, or they could die. Former smokers
could relapse, thus becoming smokers again, or remain former smokers or die. In
each cycle, individuals could have one of four comorbidities: lung cancer, coronary
heart disease, stroke and chronic obstructive pulmonary disease (COPD). The
analysis was conducted from the perspective of the UK’s NHS and the time horizon

of the analysis was lifetime. According to the model, the expected lifetime costs per
person were £12,730 for the intervention group and £12,713 for standard care. The
expected number of QALYs per person over a lifetime was estimated to be 19.7 for
the intervention group and 19.6 for the standard care group. The cost per QALY
associated with the intervention was £244, which is far below the lower NICE cost-
effectiveness threshold of £20,000. Moreover, the cost-effectiveness acceptability
analysis showed that at willingness to pay of £20,000-30,000 per additional QALY
the probability of the intervention being cost effective is 0.93-0.94. Overall, the model
was found to be robust to estimates of comorbidities, utility values, costs associated
with death and intervention costs. However, using the lower estimate of
intervention effect resulted in a cost per QALY of £150,609 and using an upper
estimate intervention was dominant. This huge variation in the results reflects the
lack of clinical evidence pertaining to smoking cessation interventions in this
population. Also, using a 10-year time frame resulted in a cost per QALY of £54,446
and the subgroup analysis indicated that the intervention was cost saving for the
female cohort. The analysis has excluded costs accruing to the PSS. However, the
authors justified this by reporting that PSS costs account for <10% of the total NHS
and social care services costs for people with psychosis and schizophrenia and so are
unlikely to affect the results. Also, a range of other costs that are relevant to the NHS
have been excluded, including psychosis and schizophrenia treatment costs and
costs of managing drug-related side effects. Moreover, the standard care definition
was adopted from the studies that were included in the meta-analysis of
intervention effect. Therefore, it is not clear if the comparator used is a good
representation of the current clinical practice in the NHS. The analysis has
incorporated the impact of smoking cessation on various comorbidities including
lung cancer, COPD, coronary heart disease and stroke. The prevalence data for
stroke and coronary heart disease were derived from a Canadian population-based
study and for COPD from a US population-based controlled study, which may be
different from prevalence rates in the UK. Similarly, EQ-5D ratings for the baseline
were from a German patient sample. Also, the treatment effect estimate was based
on a meta-analysis and authors’ assumptions, and as indicated by the sensitivity
analysis, the results are very sensitive to this estimate. The resource use data were
derived from various published sources and supplemented with authors’
assumptions. Overall this study was judged by the GDG to be partially applicable to
this guideline review and the NICE reference case, and it had potentially serious

The GDG agreed that the main aim of a smoking intervention is to either reduce or
stop smoking. Furthermore, satisfaction with services (indicating the likelihood of
continuing the intervention) and the service user’s quality of life were considered
critical outcomes. In addition to this, the GDG felt it was important to assess any
adverse effects on psychiatric symptoms as a result of smoking reduction or
cessation. Therefore, the outcomes the GDG considered to be critical were:

• anxiety and depression
• physical health
o smoking (cessation or reduction)
o weight/BMI
• quality of life

The physical harm caused by smoking is so palpable that the GDG felt it was
important to offer all people with psychosis and schizophrenia who smoke support
with smoking cessation or reduction, even if they had previously been unsuccessful
in doing so.
The GDG evaluated the evidence presented for efficacy of safety of interventions in a
schizophrenia population. Furthermore, evidence from the general population in the
NICE smoking cessation public health guideline (PH10) (NICE, 2013b) was also
considered by the GDG.
For adults with psychosis and schizophrenia who smoke, the GDG considered there
to be reasonable evidence of the benefits of bupropion for smoking cessation and
some limited evidence of its effectiveness for smoking reduction. The evidence of
smoking reduction or cessation using bupropion did not exacerbate psychosis
symptoms, or symptoms of anxiety or depression. However, the GDG was
concerned that bupropion is contraindicated in people with bipolar disorder because
of the risk of seizures and other neuropsychiatric adverse effects 16. A large number
of people with an initial diagnosis of psychosis prove to have a more specific
diagnosis of bipolar disorder. Therefore, the GDG believe that bupropion should not
be used for people with psychosis unless a diagnosis of schizophrenia is confirmed.
The GDG considered there was reasonable evidence of a benefit of varenicline for
smoking cessation for people with schizophrenia. However, there are concerns about
possible neuropsychiatric adverse effects as stated in the Summary of Product
Characteristics (SPC) 17, and found in the evidence from this review. The GDG
considered that varenicline should be prescribed cautiously for smoking cessation
for an adult with psychosis and schizophrenia, and, bearing in mind guidance from
the Royal College of Practitioners and the Royal College of Psychiatrists (Campion et
al., 2010) the service user regularly monitored for possible neuropsychiatric adverse
effects especially in the first 2-3 weeks. The GDG thought that to promote service
user choice, people should be made aware of the possible adverse effects of both
varenicline and bupropion.
16 See http://emc.medicines.org.uk/
17 See http://emc.medicines.org.uk/

There was a paucity of follow-up data evaluating the long-term efficacy of
bupropion or varenicline, however, the GDG believed that the potential negative
consequences of continuing smoking outweighed this lack of knowledge.
There was also a lack of data evaluating the efficacy of NRT in this population. The
GDG therefore considered the efficacy evidence in the general population for
smoking reduction, and the fact that there are no known contraindications (outside
of those for the general population as discussed in PH10) specifically for those with
psychosis and schizophrenia. The group decided that a transdermal nicotine patch
and other forms of NRT should also be offered to encourage smoking cessation and
reduction.
The GDG also deliberated about how best to manage smoking in inpatient settings
and judged that support should be offered to encourage those who may not want to
cease smoking completely to temporarily stop or reduce smoking by using NRT.

The health economic evidence on smoking cessation was limited to one UK study.
Despite study limitations (for instance, poor clinical evidence, the omission of
potential cost savings from reducing smoking), the results provide some evidence
that providing targeted smoking cessation interventions for adults with psychosis
and schizophrenia can be cost effective and a viable approach within the NICE
decision-making context. The positive economic finding supports the GDG view that
it is important to offer all people with psychosis and schizophrenia who smoke
support with smoking cessation.

The evidence ranged from very low to moderate quality across critical outcomes.
Reasons for downgrading included risk of bias in the included studies, high
heterogeneity and lack of precision in confidence intervals. Wide confidence
intervals were a major concern when evaluating the evidence. However, although
variance was observed in the effect size across studies, the direction of effect was
consistent across most and the small number of participants in the included trials
could have contributed to the lack of precision.
At the time of drafting this guidance, NICE public health guidance, Smoking
Cessation in Secondary Care: Acute, Maternity and Mental Health Services’ was out for
public consultation and a final post-consultation draft was not available. As of
August 2013, the public health guideline recommends varenicline or bupropion for
all people who smoke. However, the GDG thought it was of critical importance that
varenicline should only be offered to people with psychosis and schizophrenia
cautiously because of concerns about its association with an increased risk of
neuropsychiatric events. The GDG also judged it important that bupropion is not
offered to people who have a diagnosis of psychosis unless a more specific diagnosis
of schizophrenia is confirmed.

Finally, blood levels of some antipsychotics, particularly clozapine and olanzapine,
are reduced as the hydrocarbons in cigarette smoke induce the main enzyme system
responsible for the metabolism of these drugs. When smoking is stopped, enzyme
induction no longer occurs and blood levels of the affected drugs could increase to
high levels. The effect of smoking on people taking clozapine is of particular concern
and individuals can become ill unless the dose is adjusted. The GDG believes that
this should be considered in advance of smoking cessation.
7.3.8.1 Offer people with psychosis or schizophrenia who smoke help to stop
smoking, even if previous attempts have been unsuccessful. Be aware of the
potential significant impact of reducing cigarette smoking on the metabolism
of other drugs, particularly clozapine and olanzapine. [new 2014]
7.3.8.2 Consider one of the following to help people stop smoking:
• nicotine replacement therapy (usually a combination of
transdermal patches with a short-acting product such as an
inhalator, gum, lozenges or spray) for people with psychosis or
schizophrenia or
• bupropion 18 for people with a diagnosis of schizophrenia or
• varenicline for people with psychosis or schizophrenia.
Warn people taking bupropion or varenicline that there is an increased risk of
adverse neuropsychiatric symptoms and monitor them regularly, particularly
in the first 2-3 weeks. [new 2014]
7.3.8.3 For people in inpatient settings who do not want to stop smoking, offer
nicotine replacement therapy to help them to reduce or temporarily stop
smoking. [new 2014]
18 At the time of publication (February 2014), bupropion was contraindicated in people with bipolar disorder.
Therefore, it is not recommended for people with psychosis unless they have a diagnosis of schizophrenia.

8 PEER-PROVIDED AND SELF-
MANAGEMENT INTERVENTIONS
8.1 INTRODUCTION
This chapter is new for the 2014 guideline and reviews the evidence for peer-
provided interventions (see Section 8.2) and self-management interventions (see
Section 8.3). The decisions that led to the development of recommendations from
both reviews can be found in Section 8.4, and the recommendations themselves in
Section 8.5.
8.2 PEER-PROVIDED INTERVENTIONS

8.2.1 Introduction
Peer support workers have a long history as an informal element of all types of
mental health services, dating as far back as the 19th century (Basset et al., 2010).
More recently, attendees of inpatient wards and day centres have freely provided
one another with informal support, finding that contact with others with similar
experiences can bring hope and understanding. This capacity for mutual support has
been more formally harnessed through third sector and self-help agencies, for
example, Mind and the Hearing Voices Network (Hearing Voices Network, 2003),
and employing people with lived experience of substance misuse is widely accepted
in addiction services, for example, Alcoholics Anonymous. Across North America
and Australasia (Repper & Carter, 2010) peer support workers are becoming well
established within the mainstream mental health workforce, and access to such
support for people with severe mental illness has been widely advocated
internationally by service user researchers (Clay et al., 2005; Deegan, 1996; Faulkner
& Basset, 2012) and professional organisations (Bradstreet & Pratt, 2010; Halvorson
& Whitter, 2009; The Royal College of Psychiatrists Social Inclusion Scoping Group,
2009). Provision of peer support is identified as a fidelity requirement for recovery-
orientated services (Armstrong & Steffen, 2009) and commonly promoted in
literature on recovery (Scottish Recovery Network, 2005; Slade, 2009). Roles for peer
support workers have thus evolved over time, with some continuing to be informal
through peer-led groups and others developing as more intentional or formal roles.
This chapter is concerned with the latter.
One definition of peer support work is ‘social emotional support, frequently coupled
with instrumental support, that is mutually offered or provided by persons having a
mental health condition to others sharing a similar mental health condition to bring
about a desired social or personal change’ (Solomon, 2004). A key aspect of this
definition is that it is explicit about the use that is made of lived experience, or
mutuality, of mental illness. In addition, peer support should not be tokenistic (that
is, have little real commitment or understanding of the role of peers within the

system), and it should not be a way of undertaking work cheaply that would be
better done by professionals.
People who have experienced mental health problems and used services are
potentially well placed to support other service users. There is much evidence that
people with psychosis or schizophrenia find engagement with mental health services
difficult and may avoid contact (NICE, 2011). This may be because of previous bad
experiences, especially in inpatient settings, internal and external stigma,
discrimination and/or low expectations from mental health professionals about
prognosis and potential aspirations. Peers may bring experiential knowledge to help
them support others to overcome these barriers, challenge attitudes of clinical staff
and contribute to culture change within mental health services (Repper & Watson,
2012). They may also be able to credibly model recovery and coping strategies, thus
promoting hope and self-efficacy (Salzer & Shear, 2002). The opportunity to help
others may also be of therapeutic value to peers providing support (Skovholt, 1974).
Peer-provided interventions operate in a variety of ways and do not derive from a

highly specified theoretical model or have a single, well-defined goal. The critical
ingredients of peer support have been conceptualised more in terms of style and
process—for example being non-coercive, informal and focused on strengths
(Solomon, 2004)—than in terms of content. This creates challenges for the evaluation
of peer support programmes because they may differ considerably and may aim to
improve different outcomes.
Three broad types of organised peer-provided interventions have been identified

(Davidson et al., 1999):
• Mutual support groups in which relationships are reciprocal in nature, even if

some participants are viewed as more experienced or skilled than others.
• Peer-support services in which support is primarily in one direction, with one
or more clearly defined peer support worker offering support to one or more
programme participant (support is separate from or additional to standard
care provided by mental health services).
• Peer mental health service providers where people who have used mental health
services are employed by a service to provide part or all of the standard care
provided by the service.
However, even within these subtypes of peer support, programmes may vary
regarding mode of delivery (group or one to one; in person or internet-based),
duration, degree of co-location and integration with mental health services, and
content (whether highly structured and focusing on self-management or less
structured with greater focus on activity and social contact).
8.2.2 Clinical review protocol (peer-provided interventions)


guideline, can be found in Table 53 (the full review protocol and a complete list of
review questions can be found in Appendix 6; further information about the search
The review strategy was to evaluate the clinical effectiveness of interventions using
meta-analysis. However, in the absence of adequate data, the available evidence was
synthesised using narrative methods.
Table 53: Clinical review protocol for the review of peer-provided interventions
potential harms of peer-provided interventions compared with treatment as
usual or other intervention?
Sub-question (s) a. Peer support
b. Mutual support
c. Peer mental health service providers
Objectives To evaluate the clinical effectiveness of peer-provided interventions in the
treatment of psychosis and schizophrenia.
Population Included
Adults (18+) with schizophrenia (including schizophrenia-related disorders
such as schizoaffective disorder and delusional disorder) or psychosis.
Intervention(s) Peer-provided interventions
Critical outcomes • Empowerment/recovery
• Functional disability
• Quality of life
• Service use
o GP visits
o A&E visits
o Hospitalisation (admissions, days)
Electronic databases Core: CDSR, CENTRAL, DARE, Embase, HTA, MEDLINE, PreMEDLINE
SR: 1995 to June 2013
• 7-12 months’ follow-up (medium-term)
• 12 months’ follow-up (long-term)
Sub-analysis
Where data were available, sub-analyses were conducted of studies with
schizophrenia, schizoaffective disorder or psychosis.

studies.

Sixteen RCTs (N = 4,778) met the eligibility criteria for this review: BARBIC2009
(Barbic et al., 2009), CHINMAN2013 (Chinman et al., 2013), CLARKE2000 (Clarke et
al., 2000), COOK2011 (Cook et al., 2011), COOK2012 (Cook et al., 2012),
CRAIG2004A (Craig et al., 2004a), DAVIDSON2004 (Davidson, 2004),
EDMUNDSON1982 (Edmundson et al., 1982), GESTEL-TIMMERMANS2012 (Van
Gestel-Timmermans et al., 2012), KAPLAN2011 (Kaplan et al., 2011), ROGERS2007
(Rogers et al., 2007), RIVERA2007 (Rivera et al., 2007), SLEDGE2011 (Sledge et al.,
2011), SEGAL2011 (Segal et al., 2011), SELLS2006 (Sells et al., 2006), SOLOMON1995
(Solomon & Draine, 1995). All trials were published in peer-reviewed journals
between 1982 and 2012. Further information about both included and excluded
studies can be found in Appendix 15a.
For the purposes of the guideline, interventions were categorised as:

• peer support
• mutual support
• peer mental health service providers.
Of the 16 included trials, nine involved a comparison between peer-support services

and any type of control, four involved a comparison between mutual support and
any type of control, and three compared peer mental health service providers with
any control. Table 54 provides an overview of the included trials in each category.
Of the eligible trials, three included a large proportion (>75%) of participants with a
primary diagnosis of psychosis or schizophrenia. Only one of the included trials was
based in the UK/Europe.

Table 54: Study information table for trials included in the meta-analysis of peer-
provided interventions versus any alternative management strategy
Peer-support services versus Mutual-support services Peer mental health

any control versus any control service providers
versus any control
Total no. of trials k = 9; N = 2,466 k = 4; N = 2,369 k = 3; N = 411
(k); participants
(N)
Study ID BARBIC2009 EDMUNDSON1982 CLARKE2000
CHINMAN2013 KAPLAN2011 SELLS2006
COOK2011 ROGERS2007 SOLOMON1995
COOK2012 SEGAL2011
CRAIG2004A
DAVIDSON2004
GESTEL-TIMMERMANS2012
RIVERA2007
SLEDGE2011
Country Canada (k = 1) USA (k = 4) USA (k = 3)
Netherlands (k = 1)
UK (k = 1)
USA (k = 6)
Year of 2004 to 2012 1982 to 2011 1995 to 2006
publication
Mean age of 43.16 years (37.6 to 53.27 42.23 years (37 to 47 years)1 39.8 years (36.5 to 41.9
participants years) years)
(range)
Mean percentage 52.83% (20.2 to 100%) 37.9% (22.4 to 50.4%) 1 67.6% (59.5 to 82%)
of participants
with primary
diagnosis of
psychosis or
schizophrenia
(range)
Mean percentage 46.72% (11.46 to 66%) 59.9% (54 to 65.7%) 1 41.7% (38.7 to 47%)
of women (range)
Length of 8 to 52 weeks 35 to 52 weeks 52 to 104 weeks
treatment (range)
Length of follow- End of treatment only: End of treatment only: End of treatment only:
up BARBIC2009 EDMUNDSON1982 CLARKE2000
CHINMAN2013 KAPLAN2011 SELLS2006
CRAIG2004A ROGERS2007 SOLOMON1995
DAVIDSON2004 SEGAL2011
RIVERA2007
SLEDGE2011
Up to 6 months:
COOK2011
COOK2012
7-12 months:
COOK2011
Intervention type ‘Recovery Workbook’ + TAU Community network Peer-based case
(k = 1) development (k = 1) management (k = 1)
‘PEER Simpson Transfer Internet peer support email Consumer-provided
Model’ (k = 1) list (k = 1) ACT (k = 1)

‘Building Recovery of Bulletin board (k = 1) Consumer case
Individual Dreams and Goals Consumer-operated service management (k = 1)
through Education and programmes (k = 2)
Support’ (BRIDGES) + TAU (k
= 1)
‘Wellness Recovery Action
Plan’ (WRAP) + TAU (k = 1)
Peer support + TAU (k = 3)
‘The Partnership Project’ +
TAU (k = 1)
‘Recovery Is Up to You’ +
TAU (k = 1)
Comparisons TAU/usual services (k = 6) Outpatient services (k = 3) Case management (k =
Case management without Waitlist (k = 1) 2)
peer enhancement (k = 2) Professional-led ACT (k
Supported socialisation from = 1)
non-consumer (k = 1)
Note. ACT = assertive community treatment; TAU = treatment as usual.
1 EDMUNDSON1982 does not report data.

8.2.4 Clinical evidence for peer-provided interventions
Peer support
Low to very low quality evidence from up to four studies with 1,066 participants
showed that peer support had a positive effect on self-rated recovery at the end of
the intervention and at short-term follow-up. No difference was observed between
peer support and control in empowerment or quality of life at the end of treatment,
but up to two studies (N = 639) presented very low quality evidence that peer
support was more effective than control in improving these outcomes at short-term
follow-up.
Very low quality evidence from one trial with 165 participants favoured control over
peer support for the outcome of functional disability.
Three studies (N = 255) provided very low quality evidence of a beneficial effect of
peer support on contact with services at the end of the intervention. However, no
follow-up data were available. There was no conclusive evidence of any benefit of
peer support on hospitalisation or on service user satisfaction outcomes at the end of
the intervention and no follow-up data were available.

For the critical outcomes of hospitalisation, service use, satisfaction with services,
recovery and quality of life, the sub-analysis findings did not differ from the main
analysis and continued to show a benefit of peer support at the end of the
intervention. Unlike the main analysis, the sub-analysis found a large positive effect
on empowerment at the end of the intervention. However, because of a discrepancy
in the authors’ description of the empowerment measure and the data presented,
this large effect should be treated with caution.

Table 55: Summary of findings table for peer support compared with any
alternative management strategy
Patient or population: Adults with psychosis and schizophrenia

Intervention: Peer support
Comparison: Any alternative management strategy
Outcomes Illustrative comparative risks* (95% CI) Relative No. of Quality of
Assumed Corresponding risk effect participants the
risk (95% CI) (studies) evidence
Control Peer support (GRADE)
Recovery - end of N/A Mean recovery (end of treatment) N/A 1,066 ⊕⊝⊝⊝
treatment in the intervention groups was 0.24 (4 studies) Very low1,2,3
0.39 higher)
Recovery, up to 6 N/A Mean recovery (up to 6 months’ N/A 439 ⊕⊕⊝⊝
months’ follow-up follow-up) in the intervention (2 studies) Low2,3
groups was 0.23 standard
deviations higher (0.09 to 0.37
higher)
Empowerment - N/A Mean empowerment (end of N/A 286 ⊕⊝⊝⊝
end of treatment treatment) in the intervention (2 studies) Very
groups was 2.34 standard low2,3,4,5
deviations lower (7.68 lower to 3.00
higher)
Empowerment - N/A Mean empowerment (up to 6 N/A 538 ⊕⊝⊝⊝
up to 6 months’ months’ follow-up) in the (2 studies) Very low2,3,4
follow-up intervention groups was
0.25 standard deviations higher
(0.07 to 0.43 higher)
Functioning / N/A Mean functioning/disability (end N/A 165 ⊕⊝⊝⊝
disability - end of of treatment) in the intervention (1 study) Very low2,3,6
treatment groups was 0.37 standard
higher)
Quality of life - N/A Mean quality of life (end of N/A 1039 ⊕⊝⊝⊝
higher)
Quality of life- up N/A Mean quality of life (up to 6 N/A 639 ⊕⊝⊝⊝
to 6 months’ months’ follow-up) in the (2 studies) Very low2,3,4
follow-up intervention groups was 0.24
0.40 lower)
Service use, N/A Mean service use (end of N/A 255 ⊕⊝⊝⊝
contact - end of treatment) in the intervention (3 studies) Very
treatment groups was 0.22 standard low1,2,3,4
higher)
Service use, Study population RR 1.07 45 ⊕⊝⊝
hospitalisation- 429 per 459 per 1000 (0.55 to (1 study) Very low2,3,6
end of treatment 1000 (236 to 887) 2.07)
429 per 459 per 1000

1000 (236 to 888)
Satisfaction, N/A Mean satisfaction (end of N/A 332 ⊕⊝⊝⊝
questionnaire - treatment) in the intervention (3 studies) Very low2,3,4
end of treatment groups was 0.02 standard
higher)
Note. CI = confidence interval; RR = risk ratio.
footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and
the relative effect of the intervention (and its 95% CI).
6 Crucial limitation for one criterion or some limitations for multiple criteria sufficient to lower confidence in
the estimate of effect.

7 A single study of 0.00 effect.
Mutual support
presented in
Table 56. The full evidence profiles and associated forest plots can be found in
Appendix 17 and Appendix 16, respectively.
Very low quality evidence from up to three trials (N = 2,266) provided evidence
favouring mutual support for self-rated outcomes of empowerment, quality of life,
and contact with services at the end of the intervention. There was no evidence
available to assess these outcomes at follow-up. No difference was observed between
groups in hospitalisation outcomes at the end of the intervention. No data were
available for the critical outcomes of functional disability and service user
satisfaction.
Peer mental health service providers

Very low quality evidence from a single trial with 87 participants favoured control
for service user satisfaction at the end of the intervention. There was no evidence of a
difference between groups in hospitalisation at the end of the intervention. No
follow-up data were available for both outcomes and no data were available at all for
the other critical outcomes of empowerment/recovery, functional disability or
quality of life.

No difference between the sub-analysis and the main analysis was found for service
user satisfaction. No other data were available.

Table 56: Summary of findings table for mutual support compared with any

Intervention: Mutual support
Control Mutual support (GRADE)
Recovery - end of N/A Mean recovery (end of N/A 300 ⊕⊝⊝⊝
treatment treatment) in the intervention (1 study) Very low1,2,3
deviations higher (0.13 lower to
0.35 higher)
Empowerment - N/A Mean empowerment (end of N/A 2266 ⊕⊝⊝⊝
higher)
Quality of life - end N/A Mean quality of life (end of N/A 300 ⊕⊝⊝⊝
of treatment treatment) in the intervention (1 study) Very low1,3,6
higher)
Service use, Study population RR 0.63 80 ⊕⊝⊝⊝
contact - end of (0.44 to (1 study) Very low1,2,3
250 per 158 per 1000 (110 to 230)
treatment 0.92)
1000
250 per 158 per 1000 (110 to 230)
1000
hospitalisation - (0.23 to (1 study) Very low1,2,3
end of treatment 350 per 175 per 1000 (81 to 389) 1.11)
1000
350 per 175 per 1000 (81 to 389)
1000


Table 57: Summary of findings table for interventions with peer mental health
service providers compared with any alternative management strategy

Intervention: Peer mental health service providers
Control Peer mental health service (GRADE)
providers
hospitalisation - (0.45 to (1 study) Very
end of treatment 1.03) low1,2,3
544 per 370 per 1000
1000 (245 to 560)
544 per 370 per 1000
1000 (245 to 560)
Satisfaction, N/A Mean satisfaction ( end of N/A 87 ⊕⊝⊝⊝
questionnaire - end treatment) in the intervention (1 study) Very
of treatment groups was 0.48 standard low1,3,4
higher)


Overall there is inconclusive evidence concerning the efficacy for peer-provided
interventions in both magnitude and direction of the effect. When large effects are
observed, there is some concern about the validity of these findings because of the
size of the trials and variance observed across studies. Furthermore, due to the
limited evidence, no longer-term effects of the intervention can be determined.

The systematic literature search identified one economic study that assessed peer-
provided intervention for people with psychosis and schizophrenia (Lawn, 2008).
Details on the methods used for the systematic search of the economic literature are
described in Chapter 3. References to included studies and evidence tables for all
economic studies included in the guideline systematic literature review are
presented in Appendix 19. Completed methodology checklists of the studies are
provided in Appendix 18. Economic evidence profiles of studies considered during
guideline development (that is, studies that fully or partly met the applicability and

quality criteria) are presented in Appendix 17, accompanying the respective GRADE
clinical evidence profiles.
Lawn and colleagues (2008) conducted a cost analysis in Australia. The analysis was
based on a small pre- and post-observational study (n = 49). The study comprised
individuals with bipolar affective disorder, schizophrenia, schizoaffective disorder
and first episode psychosis. Standard care was defined as psychiatric inpatient care
and care by a community-based emergency team and a community mental health
team (CMHT). The analysis was conducted from the healthcare payer perspective
and considered costs of admissions, community emergency contacts and programme
provision. The authors found that peer-provided interventions led to a cost saving of
$AUD 2,308 per participant over 3 months and cost $AUD 405 to provide, resulting
in a net saving of $AUD 1,901 per participant over 3 months. The analysis was
judged to be partially applicable to this guideline review and the NICE reference
case. However, the analysis was based on a very small pre- and post-observational
study, which was prone to bias due to the inability to control for confounding
factors. Moreover, the analysis has not attempted to capture health effects and
adopted a very short time horizon that may not be sufficiently long to reflect all
important differences in costs. Also, the source of unit costs is unclear. The analysis
was therefore judged by the GDG to have very serious methodological limitations.
8.3 SELF-MANAGEMENT INTERVENTIONS

8.3.1 Introduction
Self-management refers to an ‘individual’s ability to manage the symptoms,
treatment, physical and psychosocial consequences and life style changes inherent
living with a chronic condition’ (Barlow et al., 2002). Mental illness self-management
has increased in popularity over the past decade, and programmes based on this
approach have been now widely recommended as a means of promoting recovery
and empowering service users, while simultaneously addressing service capacity
issues (Mueser et al., 2002b; Turner et al., 2008). This reflects a broader trend in
healthcare of a collaborative rather than a traditional didactic medical approach
(Mueser & Gingerich, 2011).
Objectives for self-management include: instilling hope; improving illness

management skills; providing information about the nature of the illness and
treatment options; developing strategies for self-monitoring of the illness; improving
coping strategies; and developing skills to manage life changes (Mueser & Gingerich,
2011). Training in self-management may come from mental health professionals,
peer support workers or coaches, or it may be provided partly or wholly through
information technology. The philosophical underpinning for such training in self-
management skills is one of teaching and learning, fostering active engagement and
participation. Central to this approach is also the development of individual
strategies so that self-management strategies are rooted in experience—this
approach, in turn, supports the validation of services users’ experiences, so
individuals can apply their own meaning to each topic.

Active service user participation in developing and sustaining self-management
programmes may be difficult to achieve where there is a perception of a large power
difference between mental health professionals and service users and their carers. A
relatively pessimistic view of service users’ potential has also been reported among
healthcare professionals, which may also impact on the extent to which they
promote and engage with collaborative interventions (Hansson et al., 2013). Thus,
the belief that people with psychosis or schizophrenia can contribute to their own
health management is likely to be an important condition for effective collaboration
in self-management programmes.
A number of self-management packages focused on serious mental illness have been

developed. They include the Wellness Recovery Action Plan (WRAP) (Copeland &
Mead, 2004), the Illness Management and Recovery (IMR) programme (Gingerich &
Tornvall, 2005) and the Social and Independent Living Skills (SILS) programme
(Liberman et al., 1994). Means of delivery vary widely, and may be face to face,
group-based or via written or digital materials. Professionals, carers and peers are
involved to varying extents in supported self-management programmes. Online and
other computerised self-management programmes are becoming widespread in
other areas of health, though their development for psychosis and schizophrenia has
thus far been limited. A prominent UK trend is the setting up in many areas of
recovery colleges, in which peers, carers and mental health professionals collaborate
in supporting service users in learning about mental health and recovery (Perkins et
al., 2012; Perkins & Slade, 2012). Self-management tools are a key element in this
approach. Recovery colleges are thought to provide an environment for developing
ability and knowledge on condition management and life skills. The culture and
structure of the recovery college promote responsibility and can give confidence to
‘graduates’ to access education and employment.
Several papers (Jones & Riazi, 2011; Kemp, 2011; Mueser & Gingerich, 2011) have
reviewed and summarised the elements of self-management programmes, which
include:
• psychoeducation about mental health difficulties and available treatments

and services
• relapse prevention approaches, where service users are supported in
identifying early warning signs and in developing strategies for avoiding or
attenuating the severity of relapse
• management of medication, including identification of side effects and
strategies for negotiation with professionals to optimise medication regimes
to achieve the best balance of positive and negative effects
• symptom management, including strategies for managing persistent
symptoms of psychosis, anxiety and low mood
• setting individual recovery goals and developing strategies for achieving
them

• development of life skills important for wellbeing, self-care, productivity and
leisure, for example, diet, exercise, smoking cessation, finances, safety,
relationships, organisation, home making and communication.
8.3.2 Clinical review protocol (self-management interventions)

Table 58: Clinical review protocol summary for the review of self-management
interventions
potential harms of self-management interventions compared with treatment as
usual or other intervention?
Objectives To evaluate the clinical effectiveness of self-management interventions in the
Population Included
Intervention(s) Self-management interventions
Critical outcomes • Empowerment/recovery
• Hospitalisation (admissions, days)
• Contact with secondary services
• Quality of life
• Symptoms of psychosis
o total symptoms
o positive symptoms
o negative symptoms
Electronic database Core: CDSR, CENTRAL, DARE, Embase, HTA, MEDLINE, PreMEDLINE
SR: 1995 to June 2013
Study design RCT
Sub-analysis
Where data were available, sub-analyses were conducted of studies with >75%
of the sample described as having a primary diagnosis of schizophrenia,

studies.

Twenty-five RCTs (N = 3,606) met the eligibility criteria for this review: ANZAI2002
(Anzai et al., 2002), BARBIC2009 (Barbic et al., 2009), BAUER2006 (Bauer et al., 2006),
CHAN2007 (Chan et al., 2007), COOK2011 (Cook et al., 2011), COOK2012 (Cook et
al., 2012), ECKMAN1992 (Eckman et al., 1992), FARDIG2011 (Färdig et al., 2011),
HASSON2007 (Hasson-Ohayon et al., 2007), KOPELOWICZ1998A (Kopelowicz,
1998), KOPELOWICZ1998B (Kopelowicz et al., 1998), LEVITT2009 (Levitt et al.,
2009), LIBERMAN1998 (Liberman et al., 1998), LIBERMAN2009 (Liberman &
Kopelowicz, 2009), MARDER1996 (Marder et al., 1996), NAGEL2009 (Nagel et al.,
2009), PATTERSON2003 (Patterson et al., 2003), PATTERSON2006 (Patterson et al.,
2006), SALYERS2010 (Salyers et al., 2010), SHON2002 (Shon & Park, 2002),
VREELAND2006 (Vreeland et al., 2006), WIRSHING2006 (Wirshing et al., 2006),
XIANG2006 (Xiang et al., 2006), XIANG2007 (Xiang et al., 2007), GESTEL-
TIMMERMANS2012 (Van Gestel-Timmermans et al., 2012).
All 25 trials were published in peer-reviewed journals between 1992 and 2012.
Further information about both included and excluded studies can be found in
Appendix 15a.
Of the 25 included trials, there were four evaluating the effectiveness of peer-led self-
management, and there were 21 evaluating professional-led self-management. The
GDG decided that there was not enough trial evidence to conduct separate reviews
based on these categories, therefore all trials were included in a larger review of self-
management verses any alternative management strategy.
Of the eligible trials, 18 included a large proportion (>75%) of participants with a

primary diagnosis of psychosis or schizophrenia. None of the included trials were
based in the UK and only two were based in Europe. Table 59 provides an overview
of the trials.

Table 59: Study information table for trials included in the meta-analysis of self-
management interventions versus any alternative management strategy
Self-management versus any alternative management

strategy
Total no. of trials (k); k = 25; N = 3606
participants (N)
Study ID ANZAI2002
BARBIC2009
BAUER2006
CHAN2007
COOK2011
COOK2012
ECKMAN1992
FARDIG2011
HASSON2007
KOPELOWICZ1998A
KOPELOWICZ1998B
LEVITT2009
LIBERMAN1998
LIBERMAN2009
MARDER1996
NAGEL2009
PATTERSON2003
PATTERSON2006
SALYERS2010
SHON2002
VREELAND2006
WIRSHING2006
XIANG2006
XIANG2007
Country Australia (k = 1)
Canada (k = 1)
China (k = 3)
Israel (k = 1)
Japan (k = 1)
South Korea (k = 1)
Sweden (k = 1)
USA (k = 15)
Netherlands (k = 1)
Year of publication 1992 to 2012
Mean age of participants 41.02 years (32.0 to 53.9 years)1
(Range)
Mean percentage of 79.6% (20.2 to 100%)
participants with primary
diagnosis of psychosis or
Mean percentage of women 33% (0 to 66%)
(range)
Length of treatment 1 week to 3 years.
Length of follow-up End of treatment only
BARBIC2009
BAUER2006
HASSON2007
KOPELOWICZ1998A

KOPELOWICZ1998B
MARDER1996
PATTERSON2006
SHON2002
VREELAND2006
WIRSHING2006
Up to 6 months:
COOK2011
COOK2012
NAGEL2009
PATTERSON2003
XIANG2006
XIANG2007
7-12 months:
ANZAI2002
CHAN2007
ECKMAN1992
FARDIG2011
LEVITT2009
LIBERMAN2009
NAGEL2009
>12 months:
LIBERMAN1998
LIBERMAN2009
NAGEL2009
SALYERS2010
XIANG2007
Intervention type ‘Bipolar Disorders Program’ (k = 1)
‘Transforming Relapse and Instilling Prosperity’ (TRIP) (k = 1)
‘Wellness Recovery Action Planning’ (WRAP) (k = 1)
‘Building Recovery of Individual Dreams and Goals through
Education and Support’ (BRIDGES) (k = 1)
‘Illness Management and Recovery’ (IMR) program (k = 4)
‘Social and Independent Living Skills Program’ (k = 10)
Motivational care planning + TAU (k = 1)
‘Functional Adaptation Skills Training’ (FAST) (k = 2)
Self-management education programme (k = 1)
‘Team Solutions’ (k = 1)
‘Recovery Is Up to You’ (k = 1)
‘Recovery Work Book’ (k = 1)
Comparison Occupational therapy (k = 2)
Psychoeducation (k = 1)
Supportive group therapy (k = 4)
Illness education class (k = 1)
Traditional ward occupational therapy programme (k = 1)
Group discussion (k = 1)
TAU (k = 14)
No treatment (k = 1)
Note. TAU = treatmentment as usual.
1VREELAND2006 did not report data.

8.3.4 Clinical evidence for self-management interventions
Very low quality evidence from up to ten trials (N = 1050) showed that self-
management was more effective than control in the management of positive and
negative symptoms of psychosis at the end of treatment. No difference was observed
between groups at other follow-up points in both positive and negative symptoms.
There was inconclusive evidence for the benefits of self-management on total
psychosis symptoms. No evidence of benefit was observed at the end of treatment,
but moderate quality evidence from one trial with up to 191 participants found some
benefit of self-management over control in psychotic symptoms at medium and
long-term follow-up.
Very low to moderate quality evidence from up to five trials (N = 338) showed that
self-management was more effective than control in reducing the risk of admission
in the short-term, although no difference was observed between groups at the end of
the intervention or at medium and long-term follow-up.
One study with 54 participants presented moderate quality evidence favouring self-
management in increasing contact with aftercare services.
There was no conclusive evidence of any benefit of self-management on self-rated

empowerment at the end of the intervention. However, moderate quality evidence
from one study (N = 538) provided evidence of benefit on empowerment at short-
term follow-up. Very low quality evidence from up to seven studies with 1,234
participants showed that self-management was more effective than control in
improving both self-rated and clinician-rated recovery. No difference between
groups was observed for functional disability at any follow-up point.
Low quality evidence from nine trials with 1,337 participants showed that self-
management had a positive effect on quality of life at the end of treatment. However,
at follow-up assessments, the findings were less conclusive. Low quality evidence
from up to three studies (N = 600) found no difference between groups in quality of
life at short- and long-term follow-up, but a significant difference favouring the
intervention at medium-term follow-up.
Regarding trials not included in the meta-analyses, NAGEL2009 reported the

intervention to be effective on the outcomes of interest.

For the critical outcomes of total and negative psychosis symptoms, empowerment,
hospitalisation and contact with secondary services, the sub-analysis findings did
not differ substantially from the main analysis and found no benefit of self-
management. The benefit found for quality of life was not as conclusive in sub-

analysis. Unlike the main analysis, there was no evidence of a benefit of self-
management for self-rated recovery although the findings still favoured self-
management for clinician-rated recovery. The related forest plots can be found in
Appendix 16.
Table 60: Summary of findings table for self-management compared with any

Control Self-management (GRADE)
Psychosis (total N/A Mean psychosis (total symptoms N/A 283 ⊕⊝⊝⊝
symptoms) - end of - end of treatment) in the (3 studies) Very low1,2,3
treatment intervention groups was
0.40 standard deviations lower
Psychosis (positive N/A Mean psychosis (positive N/A 1145 ⊕⊝⊝⊝
symptoms) - end of symptoms - end of treatment) in (10 studies) Very low1,3,4
treatment the intervention groups was
(0.56 lower to 0.07 lower)
Psychosis (negative N/A Mean psychosis (negative N/A 527 ⊕⊝⊝⊝
symptoms) - end of symptoms - end of treatment) in (7 studies) Very low1,3,4
(0.76 to 0.13 lower)
Psychosis (total N/A Mean psychosis (total symptoms N/A 84 ⊕⊕⊝⊝
symptoms) - up to 6 - up to 6 months’ follow-up) in (1 study) Low3,5
months’ follow-up the intervention groups was 0.23
standard deviations lower (0.66
lower to 0.2 higher)
symptoms) - up to 6 symptoms - up to 6 months’ (4 studies) Very low1,2,3
months’ follow-up follow-up) in the intervention
deviations lower (0.69 lower to
0.21 higher)
symptoms) - up to 6 symptoms - up to 6 months’ (4 studies) Very low1,2,3
months’ follow-up follow-up) in the intervention
0.22 higher)
Psychosis (total N/A Mean psychosis (total symptoms- N/A 88 ⊕⊕⊕⊕
symptoms) - 7-12 7-12 months’ follow-up) in the (1 study) High
months’ follow-up intervention groups was 1.49
to 1.01 lower)
symptoms) - 7-12 symptoms - 7-12 months’ follow- (3 studies) Very low2,3
months’ follow-up up) in the intervention groups
was 0.49 standard deviations

lower (1.28 lower to 0.3 higher)
symptoms) - 7-12 symptoms - 7-12 months’ follow- (2 studies) Very low2,3
Psychosis (total N/A Mean psychosis (total symptoms N/A 38 ⊕⊕⊕⊝
symptoms) - >12 - >12 months’ follow-up) in the (1 study) Moderate5
months’ follow-up intervention groups was 1.36
to 0.65 lower)
Psychosis (positive N/A Mean psychosis (positive N/A 141 ⊕⊕⊕⊝
symptoms) - >12 symptoms - >12 months’ follow- (2 studies) Moderate1
symptoms) - >12 symptoms - >12 months’ follow- (2 studies) Very low1,2,3
Global state - N/A Mean global state (functioning, N/A 526 ⊕⊕⊝⊝
functioning, disability - end of treatment) in (7 studies) Low1,4
disability - end of the intervention groups was 0.07
treatment standard deviations lower (0.33
Global state - N/A Mean global state (functioning, N/A 315 ⊕⊝⊝⊝
functioning, disability - up to 6 months’ (4 studies) Very low1,3,4
disability - up to 6 follow-up) in the intervention
months’ follow-up groups was 0.37 standard
0.32 higher)
Global state - N/A Mean global state (functioning, N/A 103 ⊕⊕⊝⊝
functioning, disability - 7-12 months’ follow- (1 study) Low3,5
disability - 7-12 up) in the intervention groups
months’ follow-up was 044 standard deviations
Global state - N/A Mean global state (functioning, N/A 183 ⊕⊝⊝⊝
functioning, disability - >12 months’ follow- (2 studies) Very low1,2,3
disability - >12 up) in the intervention groups
months’ follow-up was 0.56 standard deviations
Quality of life - end N/A Mean quality of life (end of N/A 1337 ⊕⊕⊝⊝
of treatment treatment) in the intervention (9 studies) Low3,4
higher)
Quality of life - up N/A Mean quality of life (up to 6 N/A 240 ⊕⊕⊝⊝
to 6 months’ follow- months’ follow-up) in the (2 studies) Low3,5
up intervention groups was 0.24
standard deviations higher (0.01
Quality of life - 7-12 N/A Mean quality of life (7-12 months’ N/A 600 ⊕⊕⊝⊝
months’ follow-up follow-up) in the intervention (3 studies) Low3,4

groups was
(0.09 to 0.60 higher)
Quality of life - >12 N/A Mean quality of life (>12 months’ N/A 118 ⊕⊕⊝⊝
months’ follow-up follow-up) in the intervention (2 studies) Low1
0.60 higher)
Empowerment - end N/A Mean empowerment (end of N/A 538 ⊕⊝⊝⊝
of treatment treatment in the intervention (3 studies) Very low1,2
groups) was 1.44 standard
2.97 higher)
Empowerment - up N/A Mean empowerment (up to 6 N/A 318 ⊕⊕⊕⊝
to 6 months’ follow- months’ follow-up) in the (1 study) Moderate
up intervention groups was 0.25
to 0.43)
Recovery (self-rated) N/A Mean recovery (self-rated - end of N/A 1234 ⊕⊝⊝⊝
- end of treatment treatment) in the intervention (7 studies) Very low1,4
deviations lower (0.49 to 0.05
lower)
Recovery (clinician- N/A Mean recovery (clinician-rated - N/A 354 ⊕⊕⊕⊝
rated) - end of end of treatment) in the (3 studies) Moderate1
(0.88 to 0.45 lower)
Recovery (self-rated) N/A Mean recovery (self-rated - up to N/A 883 ⊕⊕⊝⊝
- up to 12 months’ 12 months’ follow-up) in the (4 studies) Low1
(0.36 to 0.09 lower)
Recovery (clinician- N/A Mean recovery (clinician-rated - N/A 129 ⊕⊕⊕⊝
rated) - up to 12 up to 12 months’ follow-up) in (2 studies) Moderate1
months’ follow-up the intervention groups was 0.57
to 0.21 lower)
Service use, contact - Study population RR 0.24 54 ⊕⊕⊕⊝
end of treatment 630 per 151 per 1000 (0.09 to (1 study) Moderate5
1000 (57 to 384) 0.61)
Service use - N/A The mean service use N/A 122 ⊕⊕⊕⊝
hospitalisation - end (hospitalisation, end of treatment (1 study) Moderate5
of treatment - days - days hospitalised) in the
hospitalised intervention groups was
Service use - Study population RR 1.06 122 ⊕⊕⊝⊝
hospitalisation - end 288 per 305 per 1000 (0.61 to (1 study) Low1
of treatment 1000 (175 to 532) 1.85)
Service use - Study population RR 0.23 269 ⊕⊕⊕⊝
hospitalisation - up 118 per 27 per 1000 (0.08 to (3 studies) Moderate5
to 6 months’ follow- 0.7)

up 1000 (9 to 82)
Service use - Study population RR 0.77 238 ⊕⊕⊝⊝
hospitalisation - 7- 181 per 139 per 1000 (0.43 to (3 studies) Low1
12 months’ follow- 1000 (78 to 252) 1.39)
up
Service use - Study population RR 0.66 338 ⊕⊝⊝⊝
hospitalisation - >12 192 per 127 per 1000 (0.23 to (4 studies) Very low1,4
months’ follow-up 1000 (44 to 369) 1.92)
Service use - N/A Mean service use (hospitalisation N/A 122 ⊕⊕⊕⊝
hospitalisation - >12 - >12 months’ follow-up - days (1 study) Moderate5
months’ follow-up - hospitalised) in the intervention
days hospitalised groups was 0.15 standard
0.51 higher)

Overall, the evidence suggests that self-management interventions are effective for
reducing symptoms of psychosis. However, this benefit was less conclusive for
reducing the risk of hospitalisation. Self-management was effective at improving
quality of life at the end of the intervention, with some less certain evidence of long-
term benefit. Self-management was also found to be beneficial for aiding recovery in
both self-and clinician-rated outcomes. This effect was sustained at long-term
follow-up. There was no conclusive evidence of a beneficial effect of self-
management on functional disability.

No studies assessing the cost effectiveness of self-management interventions for
adults with psychosis and schizophrenia were identified by the systematic search of
the economic literature undertaken for this guideline. Details on the methods used
for the systematic search of the economic literature are described in Chapter 3.

The GDG judged that the aim of peer-provided and self-management interventions
were to manage symptoms and thus reduce the risk of hospitalisation because of
relapse. The GDG also thought that self-management interventions aimed to
empower the service user and improve quality of life and day–to-day functioning.
Therefore, the GDG decided that the critical outcomes were:
For self-management:
• empowerment/recovery
• functional disability
• quality of life
• hospitalisation (admissions, days)
• contact with secondary services
• symptoms of psychosis

o total symptoms
o positive symptoms
o negative symptoms.
For peer-provided interventions:

• empowerment/ recovery
• functional disability
• quality of life
• service use
o GP visits
o A&E visits
o hospitalisation (admissions, days)

The GDG considered the benefits of peer-provided interventions and self-
management for symptom management. Although there was some evidence of
improvement in symptoms at the end of the intervention for self-management (not
for peer-provided interventions), data were limited at any further follow-up point.
The GDG thought that self-management and peer support were likely to be
beneficial for people with psychosis and schizophrenia, but should not be provided
as the sole intervention because they were not designed as stand-alone treatments.
However, the GDG considered that both interventions should be provided as
additional support for people throughout all phases of the illness.

There was only one economic study that attempted to assess the cost savings
associated with peer-provided interventions for adults with psychosis and
schizophrenia; however the GDG judged it to have very serious limitations. No
studies assessing the cost effectiveness of self-management interventions for adults
with psychosis and schizophrenia were identified by the systematic review of the
economic literature. Due to the lack of clinical data it was decided that formal
economic modelling of peer-provided or self-management interventions in this area
would not be useful in decision-making. Nevertheless, the GDG judged that the
costs of providing such interventions are justified by the expected clinical benefits,
that is, aiding recovery in both self- and clinician-rated outcomes. Moreover, it is
likely that the costs of providing such interventions will be offset, at least partially,
by cost savings in health services resulting from improvements in symptoms of
psychosis.

For both peer-provided and self-management interventions, the quality of the
evidence ranged from very low to high. The evidence for peer support was of
particular poor quality and ranged from very low to low across critical outcomes.
Reasons for downgrading concerned risk of bias, high heterogeneity or lack of
precision in confidence intervals, which crossed clinical decision thresholds.

Heterogeneity was a major concern when evaluating the evidence. However,
although variance was observed in the effect size across studies, the direction of
effect was consistent across most studies. Furthermore, wide confidence intervals
were also of concern to the GDG. This problem was particularly found for outcomes
with low numbers of included studies and participants. The GDG considered these
quality issues when discussing possible recommendations.
The GDG considered it important to define the components of peer support and self-
management interventions. The components included in the reviews were generally
well specified and therefore the GDG used this information as a basis of discussion
when developing a recommendation.
8.5 RECOMMENDATIONS
8.5.1.1 Consider peer support for people with psychosis or schizophrenia to help
improve service user experience and quality of life. Peer support should be
delivered by a trained peer support worker who has recovered from
psychosis or schizophrenia and remains stable. Peer support workers should
receive support from their whole team, and support and mentorship from
experienced peer workers. [new 2014]
8.5.1.2 Consider a manualised self-management programme delivered face-to-face
with service users, as part of the treatment and management of psychosis or
schizophrenia. [new 2014]
8.5.1.3 Peer support and self-management programmes should include information
and advice about:
• psychosis and schizophrenia
• effective use of medication
• identifying and managing symptoms
• accessing mental health and other support services
• coping with stress and other problems
• what to do in a crisis
• building a social support network
• preventing relapse and setting personal recovery goals. [new 2014]

8.5.2.1 What is the clinical and cost effectiveness of peer support interventions in
people with psychosis and schizophrenia? (see Appendix 10 for further
details) [2014]

9 PSYCHOLOGICAL THERAPY AND
PSYCHOSOCIAL INTERVENTIONS
This chapter has been partially updated for the 2014 guideline. Most sections remain
unchanged from the 2009 guideline, however some of the recommendations have
been updated to bring them in line with the recommendations from Psychosis and
Schizophrenia in Children and Young People. This was considered necessary to avoid
discrepancies between the child and adult guidelines, particularly regarding early
intervention. Consequently new sections have been added to the evidence to
recommendations section. In addition some recommendations from the 2009
guideline have been amended to improve the wording and structure with no
important changes to the context and meaning of the recommendation. In addition, a
new review was conducted for the psychological management of trauma (section
9.11) because of the inclusion of people with psychosis for this update and the
association of trauma with the development of psychosis.
Sections of the guideline where the evidence has not been updated since 2002 are
marked as **2002**_**2002** and where the evidence has not be updated since 2009,
marked by asterisks (**2009**_**2009**).
Please note that all references to study IDs in sections that have not been updated in
this chapter can be found in Appendix 22c.
9.1 INTRODUCTION
** 2009**Psychological therapies and psychosocial interventions in the treatment of
schizophrenia have gained momentum over the past 3 decades. This can be
attributed to at least two main factors. First, there has been growing recognition of
the importance of psychological processes in psychosis, both as contributors to onset
and persistence, and in terms of the negative psychological impact of a diagnosis of
schizophrenia on the individual’s well-being, psychosocial functioning and life
opportunities. Psychological and psychosocial interventions for psychosis have been
developed to address these needs. Second, although pharmacological interventions
have been the mainstay of treatment since their introduction in the 1950s, they have
a number of limitations. These include limited response of some people to
antipsychotic medication, high incidence of disabling side effects and poor
adherence to treatment. Recognition of these limitations has paved the way for
acceptance of a more broadly-based approach, combining different treatment
options tailored to the needs of individual service users and their families. Such
treatment options include psychological therapies and psychosocial interventions.
Recently, emphasis has also been placed on the value of multidisciplinary
formulation and reflective practice, particularly where psychologists and allied

mental health professionals operate within multidisciplinary teams (British
Psychological Society, 2007).
The ‘New Ways of Working’ report (British Psychological Society, 2007) details the
increasing demand by both service users and carers to gain access to psychological
interventions, and the increasing recognition of these interventions in the treatment
and management of serious mental illnesses including schizophrenia. The report
proposes that a large expansion of training of psychologists and psychological
therapists is needed to increase the workforce competent in the provision of
psychological therapies. This chapter addresses the evidence base for the application
of psychological and psychosocial treatments, generally in combination with
antipsychotic medication, in the treatment of schizophrenia, for individuals, groups
and families.
9.1.1 The stress-vulnerability model

Although the rationales for medical, psychological and psychosocial interventions
are derived from a variety of different biological, psychological and social theories,
the development of the stress-vulnerability model (Nuechterlein, 1987; Zubin &
Spring, 1977) has undoubtedly facilitated the theoretical and practical integration of
disparate treatment approaches (see Chapter 2). In this model, individuals develop
vulnerability to psychosis attributable to biological, psychological and/or social
factors; treatments, whether pharmacological or psychological, then aim to protect a
vulnerable individual and reduce the likelihood of relapse, reduce the severity of the
psychotic episode and treat the problems associated with persisting symptoms.
Psychological interventions may, in addition, aim to improve specific psychological
or social aspects of functioning and to have a longer-term effect upon an individual’s
vulnerability.
9.1.2 Engagement
A prerequisite for any psychological or other treatment is the effective engagement
of the service user in a positive therapeutic or treatment alliance (Roth et al., 1996).
Engaging people effectively during an acute schizophrenic illness is often difficult
and demands considerable flexibility in the approach and pace of therapeutic
working. Moreover, once engaged in a positive therapeutic alliance, it is equally
necessary to maintain this relationship, often over long periods, with the added
problem that such an alliance may wax and wane, especially in the event of service
users becoming subject to compulsory treatment under the Mental Health Act.
Special challenges in the treatment of schizophrenia include social withdrawal,
cognitive and information-processing problems, developing a shared view with the
service user about the nature of the illness, and the impact of stigma and social
exclusion.
9.1.3 Aims of psychological therapy and psychosocial interventions

The aims of psychological and psychosocial interventions in the treatment of a
person with schizophrenia are numerous. Particular treatments may be intended to

improve one or more of the following outcomes: to decrease the person’s
vulnerability; reduce the impact of stressful events and situations; decrease distress
and disability; minimise symptoms; improve quality of life; reduce risk; improve
communication and coping skills; and/or enhance treatment adherence. As far as
possible, research into psychological interventions needs to address a wide range of
outcomes.
9.1.4 Therapeutic approaches identified

The following psychological therapies and psychosocial interventions were
reviewed:
• adherence therapy
• arts therapies
• cognitive behavioural therapy
• cognitive remediation
• counselling and supportive therapy
• family intervention
• psychodynamic and psychoanalytic therapies
• psychoeducation
• social skills training**2009**
• psychological management of trauma.
**2009** The primary clinical questions addressed in this chapter can be found in Box
1.
Box 1: Primary clinical questions addressed in this chapter
Initial treatment
For people with first-episode or early schizophrenia, what are the benefits and
downsides of psychological/psychosocial interventions when compared with
alternative management strategies at initiation of treatment?
Acute treatment
For people with an acute exacerbation or recurrence of schizophrenia, what are the
benefits and downsides of psychological/psychosocial interventions when
compared with alternative management strategies?
Promoting recovery in people with schizophrenia that is in remission
For people with schizophrenia that is in remission, what are the benefits and
downsides of psychological/psychosocial interventions when compared with
alternative management strategies?
Promoting recovery in people with schizophrenia who have had an inadequate or no

response to treatment

For people with schizophrenia who have an inadequate or no response to
treatment, what are the benefits and downsides of psychological/ psychosocial
interventions when compared with alternative management strategies?**2009**
Psychological management of trauma

For adults with psychosis and schizophrenia, what are the benefits and/or
potential harms of psychological management strategies for previous trauma
compared with treatment as usual or another intervention?
9.1.5 Multi-modal interventions

**2009** Some researchers have combined two psychological and/or psychosocial
interventions to attempt to increase the effectiveness of the intervention. For
example, a course of family intervention may be combined with a module of social
skills training. The combinations are various and thus these multi-modal
interventions do not form a homogenous group of interventions that can be analysed
together. Therefore, multi-modal interventions that combined psychological and
psychosocial treatments within the scope of this review were included in the
primary analysis for each intervention review. Sensitivity analyses were conducted
to test the effect, if any, of removing these multi-modal interventions. Where papers
reported more than two treatment arms (for example, family intervention only
versus social skills training only versus family intervention plus social skills
training), only data from the single intervention arms was entered into the
appropriate analysis (for example, family intervention only versus social skills
training only). Papers assessing the efficacy of psychological treatments as adjuncts
to discrete treatments outside the scope of the 2009 guideline (for example,
supported employment and pre-vocational training) were excluded from the
analysis.
It is, however, worth noting that although some of the papers included in the 2002
guideline can be classed as multi-modal treatments because they systematically
combine elements such as, for example, family intervention, social skills training and
CBT, this needs to be understood in the context of the standard care available at the
time. In particular, there has been a recent emphasis on incorporating active
elements, particularly psychoeducation, into a more comprehensive package of
standard care. Elements included in the experimental arms of older studies may now
be considered routine elements of good standard care. It should also be noted that
standard care differs across countries.
Definition
To be classified as multi-modal, an intervention needed to be composed of the
following:
• a treatment programme where two or more specific psychological
interventions (as defined above) were combined in a systematic and
programmed way; and

• the intervention was conducted with the specific intention of producing a
benefit over and above that which might be achieved by a single intervention
alone.
In addition, multi-modal treatments could provide specific interventions,
either concurrently or consecutively.
9.1.6 Competence to deliver psychological therapies

For the purpose of implementing the guideline, it is important to have an
understanding of the therapists’ level of competence in the psychological therapy
trials that were included. Each of the psychological therapy papers was reviewed for
details of training or level of competence of the therapists delivering the
intervention 21.
9.2 ADHERENCE THERAPY

9.2.1 Introduction
Pharmacological interventions have been the mainstay of treatment since their
introduction in the 1950s; however, about 50% of people with schizophrenia and
schizophreniform disorder are believed to be non-adherent to (or non-compliant
with) their medication (Nose et al., 2003). It is estimated that non-adherence to
medication leads to a higher relapse rate, repeated hospital admissions, and
therefore increased economic and social burden for the service users themselves as
well as for mental health services (Gray et al., 2006; Robinson et al., 1999).
Against this background, ‘compliance therapy’ was first developed by Kemp and
colleagues (1996; 1998) to target service users with schizophrenia and psychosis. The
therapy aims to improve service users’ attitude to medication and treatment
adherence, and thus hypothetically enhance their clinical outcomes, and prevent
potential and future relapse (Kemp et al., 1996; Kemp et al., 1998). Recently, the
terms ‘adherence’ and ‘concordance’ have been used synonymously to denote
‘compliance therapy’ and its major aim (that is, adherence to medication), as
reflected in emerging literature (McIntosh et al., 2006). Overall, ‘adherence therapy’
is the commonly accepted term used contemporarily.
Adherence therapy is designed as a brief and pragmatic intervention, borrowing

techniques and principles from motivational interviewing (Miller & Rollnick, 1991),
psychoeducation and cognitive therapy (Kemp et al., 1996). A typical adherence
therapy course offered to a service user with psychosis usually comprises four to
eight sessions, each lasting from roughly 30 minutes to 1 hour (Gray et al., 2006;
Kemp et al., 1996). The intervention uses a phased approach to:
• assess and review the service user’s illness and medication history
• explore his or her ambivalence to treatment, maintenance medication and
stigma
21Training and competency reviews are presented only for recommended interventions.

• conduct a medication problem-solving exercise to establish the service user’s
attitude to future medication use.
Definition
Adherence therapy was defined as:
• any programme involving interaction between service provider and service
user, during which service users are provided with support, information and
management strategies to improve their adherence to medication and/or with
the specific aim of improving symptoms, quality of life and preventing
relapse.
To be considered as well defined, the strategy should be tailored to the needs of

individuals.

The review protocol, including information about the databases searched and the
eligibility criteria can be found in Table 61. The primary clinical questions can be
found in Appendix 21. A new systematic search for relevant studies was conducted
for the 2009 guideline. The search identified an existing Cochrane review (McIntosh
et al., 2006) which was used to identify papers prior to 2002 (further information
about the search strategy can be found in Appendix 20).
Table 61: Clinical review protocol for the review of adherence therapy
Electronic databases CINAHL, CENTRAL, EMBASE, MEDLINE, PsycINFO
Date searched 1 January 2002 to 30 July 2008

Study design RCT (≥10 participants per arm)
Patient population Adults (18+) with schizophrenia (including schizophrenia-related
disorders)
Excluded populations Very late onset schizophrenia (onset after age 60) Other psychotic
disorders, such as bipolar disorder, mania or depressive psychosis
People with coexisting learning difficulties, significant physical or sensory
difficulties, or substance misuse
Interventions Adherence therapy

Comparator Any alternative management strategy
Critical outcomes Mortality (suicide)

Global state (relapse, rehospitalisation)
Mental state (total symptoms, depression)
Psychosocial functioning
Adherence to antipsychotic treatment
Insight
Quality of life
Leaving the study early for any reason
Adverse events

9.2.3 Studies considered for review22
Five RCTs (N = 649) met the inclusion criteria for the 2009 guideline review.
Although broadly based on a cognitive behavioural approach, KEMP1996 was
reclassified as an adherence therapy paper because the primary aim of the
intervention was to improve adherence and attitudes towards medication. All of the
trials were published in peer-reviewed journals between 1996 and 2007. In addition,
two studies were excluded from the analysis because they failed to meet the
intervention definition (further information about both included and excluded
studies can be found in Appendix 22c).
9.2.4 Adherence therapy versus control

For the 2009 guideline, five RCTs of adherence therapy versus any type of control
were included in the meta-analysis (see Table 62 for a summary of the study
characteristics). Forest plots and/or data tables for each outcome can be found in
Appendix 23d.

The limited evidence from KEMP1996 regarding improvements in measures of
compliance and insight has not been supported by new studies, including those with
follow-up measures. Although there is limited and inconsistent evidence of
improved attitudes towards medication, adherence therapy did not have an effect on
symptoms, quality of life, relapse or rehospitalisation.
9.2.6 Health economic evidence

The systematic search of the economic literature identified one study that assessed
the cost effectiveness of adherence therapy for people with acute psychosis treated in
an inpatient setting in the UK (Healey et al., 1998). The study was conducted
alongside the RCT described in KEMP1996. The comparator of adherence therapy
was supportive counselling. The study sample consisted of 74 people with
schizophrenia, affective disorders with psychotic features or schizoaffective disorder
who were hospitalised for psychosis. The time horizon of the economic analysis was
18 months (RCT period plus naturalistic follow-up). Costs consisted of those to the
NHS (inpatient, outpatient, day-hospital care, accident and emergency services,
primary and community care) and criminal justice system costs incurred by arrests,
court appearances, probation, and so on. Outcomes included relapse rates, BPRS and
GAF scores, Drug Attitude Inventory (DAI) scores, Insight scale scores and levels of
compliance with antipsychotic medication. Adherence therapy was reported to have
a significant positive effect over supportive counselling in terms of relapse, GAF,
DAI and Insight scale scores as well as compliance at various follow-up time points.
The two interventions were associated with similar costs: mean weekly cost per
person over 18 months was £175 for adherence therapy and £193 for supportive
22Here and elsewhere in this chapter, each study considered for review is referred to by a study ID, with studies
included in the previous guideline in lower case and new studies in upper case (primary author and date).
References for included studies denoted by study IDs can be found in Appendix 22c.

counselling in 1995/96 prices (p = 0.92). Because of high rates of attrition, the sample
size at endpoint (N = 46) was adequate to detect a 30% difference in costs at the 5%
level of significance. The authors suggested that adherence therapy was a cost-
effective intervention in the UK because it was more effective than supportive
counselling at a similar cost.
Table 62: Summary of study characteristics for adherence therapy
Adherence therapy versus any control

K (total N) 5 (649)
Study ID GRAY2006
KEMP1996
MANEESAKORN2007
ODONNELL2003
TSANG2005
Diagnosis 58–100% schizophrenia or other related diagnoses (DSM-III or IV)
Baseline severity BPRS total:
Mean (SD)~45 (13) GRAY2006
Mean (SD)~58 (14) KEMP1996
Mean (SD)~69 (20) ODONNELL2003
Mean (SD)~44 (8) TSANG2005
PANSS total:
Mean (SD)~59 (13) MANEESAKORN2007
Number of sessions Range: 4–8
Length of treatment Range: Maximum 3–20 weeks (GRAY2006, KEMP1996;
MANEESAKORN2007)
Length of follow-up Up to 12 months: GRAY2006, ODONNEL2003, TSANG2005
Up to 18 months: KEMP1996
Setting Inpatient: KEMP1996, MANEESAKORN2007, ODONNELL2003,
TSANG2005
Inpatient and outpatient: GRAY2006
Details on the methods used for the systematic search of the economic literature are
described in Appendix 28 . References to included/excluded studies and evidence
tables for all economic studies included in the guideline systematic literature review
are presented in the form of evidence tables in Appendix 25.

The 2009 guideline review found no consistent evidence to suggest that adherence
therapy is effective in improving the critical outcomes of schizophrenia when
compared with any other control. Although one UK-based study (KEMP1996)
reported positive results for measures of adherence and drug attitudes, these
findings have not been supported in recent, larger-scale investigations. It is also
noteworthy that a proportion of participants in the KEMP1996 study had a primary
diagnosis of a mood disorder and that, in an 18-month follow-up paper, the authors
stated that ‘subgroup analyses revealed the following: patients with schizophrenia

tended to have a less favourable outcome in terms of social functioning, symptom
level, insight and treatment attitudes’.
One economic analysis, conducted alongside KEMP1996, suggested that adherence

therapy could be a cost-effective option for people experiencing acute psychosis in
the UK because it was more effective than its comparator at a similar total cost. In
addition to the aforementioned limitations of the KEMP1996 study, because of high
attrition rates the sample was very small, making it difficult to establish such a
hypothesis.
Based on the limited health economic evidence and lack of clinical effectiveness, the
GDG therefore concluded that there is no robust evidence for the use of adherence
therapy as a discrete intervention.
9.2.8.1 Do not offer adherence therapy (as a specific intervention) to people with
psychosis or schizophrenia. [2009]
9.3 ARTS THERAPIES

9.3.1 Introduction
The arts therapy professions in the US and Europe have their roots in late 19th and
early 20th century hospitals, where involvement in the arts was used by patients and
interested clinicians as a potential aid to recovery. This became more prevalent after
the influx of war veterans in the 1940s, which led to the emergence of formal training
and professional bodies for art, music, drama and dance movement therapies. These
treatments were further developed in psychiatric settings in the latter half of the
20th century (Bunt, 1994; Wood, 1997).
While the four modalities use a variety of techniques and arts media, all focus on the
creation of a working therapeutic relationship in which strong emotions can be
expressed and processed. The art form is also seen as a safe way to experiment with
relating to others in a meaningful way when words can be difficult. A variety of
psychotherapeutic theories are used to understand the interactions between
patient(s) and therapist but psychodynamic models (see Section9.8) tend to
predominate in the UK (Crawford & Patterson, 2007).
More recently, approaches to working with people with psychosis using arts
therapies have begun to be more clearly defined, taking into consideration the phase
and symptomatology of the illness (Gilroy & McNeilly, 2000; Jones, 1996). The arts
therapies described in the studies included in this review have predominantly
emphasised expression, communication, social connection and self-awareness
through supportive and interactive experiences, with less emphasis on the use of
‘uncovering’ psychoanalytic approaches (Green et al., 1987; Rohricht & Priebe,
2006; Talwar et al., 2006; Ulrich et al., 2007; Yang et al., 1998).

Art, music, drama and dance movement therapists 23 practising in the UK are state
registered, regulated by the Health Professions Council, which requires specialist
training at Master’s level.
Definition
Arts therapies are complex interventions that combine psychotherapeutic techniques
with activities aimed at promoting creative expression. In all arts therapies:
• the creative process is used to facilitate self-expression within a specific
therapeutic framework
• the aesthetic form is used to ‘contain’ and give meaning to the service user’s
experience
• the artistic medium is used as a bridge to verbal dialogue and insight-based
psychological development if appropriate
• the aim is to enable the patient to experience him/herself differently and
develop new ways of relating to others.
Arts therapies currently provided in the UK comprise: art therapy or art

psychotherapy, dance movement therapy, body psychotherapy, drama therapy and
music therapy.

eligibility criteria, can be found in Table 63. The primary clinical questions can be
found in Box 1 (further information about the search strategy can be found in
Appendix 20).
Table 63: Clinical review protocol for the review of arts therapies

Date searched Database inception to 30 July 2008
Study design RCT (≥ 10 participants per arm)

disorders)
Excluded populations Very late onset schizophrenia (onset after age 60) Other psychotic
Interventions Arts therapies

23Registration pending.

Quality of life
Adverse events

Seven RCTs (N = 406) met the inclusion criteria for the 2009 guideline review. All
trials were published in peer-reviewed journals between 1974 and 2007 (further
information about both included and excluded studies can be found in Appendix
22c).
9.3.4 Arts therapies versus any control

For the 2009 guideline review, six out of the seven RCTs were included in the meta-
analysis of arts therapies versus any type of control (see Table 64 for a summary of
the study characteristics). One of the included studies (NITSUN1974) did not
provide any useable data for any of the critical outcomes listed in the review
protocol. Sub-analyses were used to examine treatment modality and setting. Forest
plots and/or data tables for each outcome can be found in Appendix 23d.
Table 64: Summary of study characteristics for arts therapies
Arts therapies versus any control

K (totalN) 6 (382)
StudyID GREEN1987
RICHARDSON2007
ROHRICHT2006
TALWAR2006
ULRICH2007
YANG1998
Diagnosis 50–100%schizophrenia or other related diagnoses
(DSM-III or IV)
Baseline severity BPRS total:
Mean (SD): ~16 ( 9) RICHARDSON2007
Mean (SD): ~40 (8) YANG1998
PANSStotal:
Mean (SD): ~78 (18) ROHRICHT2006
Mean (SD): ~72 (13) TALWAR2006
Treatment modality Art: GREEN1987, RICHARDSON2007
Body-orientated: ROHRICHT2006
Music: TALWAR2006, ULRICH2007,YANG1998
Length of treatment Range: 5–20 weeks
Length of follow-up Up to 6 months: RICHARDSON2007, ROHRICHT2006

Setting Inpatient: TALWAR2006, ULRICH2007,YANG1998
Outpatient: GREEN1987, RICHARDSON2007, ROHRICHT2006

The review found consistent evidence that arts therapies are effective in reducing
negative symptoms when compared with any other control. There was some
evidence indicating that the medium to large effects found at the end of treatment
were sustained at up to 6 months’ follow-up. Additionally, there is consistent
evidence to indicate a medium effect size regardless of the modality used within the
intervention (that is, music, body-orientated or art), and that arts therapies were
equally as effective in reducing negative symptoms in both inpatient and outpatient
populations.
9.3.6 Health economic considerations

No evidence on the cost effectiveness of arts therapies for people with schizophrenia
was identified by the systematic search of the economic literature. Details on the
Appendix 11.
The clinical studies on arts therapies included in the guideline systematic literature
review described interventions consisting of 12 sessions on average. These
programmes are usually delivered by one therapist to groups of six to eight people
in the UK and have an average duration of 1 hour.
Arts therapies are provided by therapists with a specialist training at Master’s level.
The unit cost of a therapist providing arts therapies was not available. The salary
scale of an arts therapist lies across bands 7 and 8a, which is comparable to the salary
level of a clinical psychologist. The unit cost of a clinical psychologist is
£67 per hour of client contact in 2006/07 prices (Curtis, 2007). This estimate has been
based on the mid-point of Agenda for Change salaries band 7 of the April 2006 pay
scale according to the National Profile for Clinical Psychologists, Counsellors and
Psychotherapists (NHS Employers, 2006). It includes salary, salary oncosts,
overheads and capital overheads, but does not take into account qualification costs
because the latter are not available for clinical psychologists.
Based on the estimated staff time associated with an arts therapy programme (as
described above) and the unit cost of a clinical psychologist, the average cost of arts
therapy per person participating in such a programme would range between £100
and £135 in 2006/07 prices.
Using the lower cost-effectiveness threshold of £20,000 per QALY set by NICE
(NICE, 2008b), a simple threshold analysis indicated that arts therapies are cost
effective if they improve the HRQoL of people with schizophrenia by 0.005 to 0.007
annually, on a scale of 0 (death) to 1 (perfect health). Using the upper cost-

effectiveness threshold of £30,000 per QALY, the improvement in HRQoL of people
in schizophrenia required for arts therapies to be cost effective fell by 0.003 to 0.004
annually.

The clinical review indicated that arts therapies are effective in reducing negative
symptoms across a range of treatment modalities, and for both inpatient and
outpatient populations. The majority of trials included in the review utilised a
group-based approach. It is noteworthy that in all of the UK-based studies the
therapists conducting the intervention were all Health Professions Council (HPC)
trained and accredited, with the equivalent level of training occurring in the non-UK
based studies.
The cost of arts therapies was estimated at roughly £100 to £135 per person with
schizophrenia (2006/07 prices); a simple threshold analysis showed that if arts
therapies improved the HRQoL of people with schizophrenia by approximately
0.006 annually (on a scale of 0 to 1) then they would be cost effective, according to
the lower NICE cost-effectiveness threshold. Using the upper NICE cost-
effectiveness threshold, improvement in HRQoL would need to approximate 0.0035
annually for the intervention to be considered cost effective. Use of this upper cost-
effectiveness threshold can be justified because arts therapies are the only
interventions demonstrated to have medium to large effects on negative symptoms
in people with schizophrenia. The GDG estimated that the magnitude of the
improvement in negative symptoms associated with arts therapies (SMD -0.59 with
95% CIs -0.83 to -0.36) could be translated into an improvement in HRQoL probably
above 0.0035, and possibly even above 0.006 annually, given that the therapeutic
effect of arts therapies was shown to last (and was even enhanced) at least up to 6
months following treatment (SMD -0.77 with 95% CIs -1.27 to -0.26).
At present, the data for the effectiveness of arts therapies on other outcomes, such as
social functioning and quality of life, is still very limited and infrequently reported in
trials. Consequently, the GDG recommends that further large-scale investigations of
arts therapies should be undertaken to increase the current evidence base. Despite
this small but emerging evidence base, the GDG recognise that arts therapies are
currently the only interventions (both psychological and pharmacological) to
demonstrate consistent efficacy in the reduction of negative symptoms. This, taken
in combination with the economic analysis, has led to the following
recommendations.
Subsequent acute episodes
9.3.8.1 Consider offering arts therapies to all people with psychosis or
schizophrenia, particularly for the alleviation of negative symptoms. This
can be started either during the acute phase or later, including in inpatient
settings. [2009]

9.3.8.2 Arts therapies should be provided by a Health and Care Professions Council
registered arts therapist with previous experience of working with people
with psychosis or schizophrenia. The intervention should be provided in
groups unless difficulties with acceptability and access and engagement
indicate otherwise. Arts therapies should combine psychotherapeutic
techniques with activity aimed at promoting creative expression, which is
often unstructured and led by the service user. Aims of arts therapies should
include:
• enabling people with psychosis or schizophrenia to experience
themselves differently and to develop new ways of relating to
others
• helping people to express themselves and to organise their
experience into a satisfying aesthetic form
• helping people to accept and understand feelings that may have
emerged during the creative process (including, in some cases, how
they came to have these feelings) at a pace suited to the person.
[2009]
9.3.8.3 When psychological treatments, including arts therapies, are started in the
acute phase (including in inpatient settings), the full course should be
continued after discharge without unnecessary interruption. [2009]
Promoting recovery
9.3.8.4 Consider offering arts therapies to assist in promoting recovery, particularly
in people with negative symptoms. [2009]

9.3.9.1 An adequately powered RCT should be conducted to investigate the clinical
and cost effectiveness of arts therapies compared with an active control (for
example, sham music therapy) in people with schizophrenia.[2009]
9.3.9.2 An adequately powered RCT should be conducted to investigate the most
appropriate duration and number of sessions for arts therapies in people
with schizophrenia.[2009]
9.4 COGNITIVE BEHAVIOURAL THERAPY

9.4.1 Introduction
CBT is based on the premise that there is a relationship between thoughts, feelings
and behaviour. Although Albert Ellis first developed CBT (which he called rational
emotive behaviour therapy) in the 1960s, most CBT practiced in the present day has
its origins in the work of Aaron T. Beck. Beck developed CBT for the treatment of
depression in the 1970s (Beck, 1979), but since then it has been found to be an
effective treatment in a wide range of mental health problems including anxiety
disorders, obsessive-compulsive disorder, bulimia nervosa and PTSD. In the early
1990s, following an increased understanding of the cognitive psychology of

psychotic symptoms (Frith, 1992; Garety & Hemsley, 1994; Slade & Bentall, 1988),
interest grew in the application of CBT for people with psychotic disorders. Early
CBT trials tended to be particularly symptom focused, helping service users develop
coping strategies to manage hallucinations (Tarrier et al., 1993). Since then, however,
CBT for psychosis (CBTp) has evolved and now tends to be formulation based.
As with other psychological interventions, CBT depends upon the effective

development of a positive therapeutic alliance (Roth et al., 1996). On the whole, the
aim is to help the individual normalise and make sense of their psychotic
experiences, and to reduce the associated distress and impact on functioning. CBTp
trials have investigated a range of outcomes over the years; these include symptom
reduction (positive, negative and general symptoms) (Rector et al., 2003), relapse
reduction (Garety et al., 2008), social functioning (Startup et al., 2004), and insight
(Turkington et al., 2002). More recently, researchers have shown an interest in the
impact of CBTp beyond the sole reduction of psychotic phenomena and are looking
at changes in distress and problematic behaviour associated with these experiences
(Trower et al., 2004). Furthermore, the populations targeted have expanded, with
recent developments in CBTp focusing on the treatment of first episode psychosis
(Jackson et al., 2005; Jackson et al., 2008), and people with schizophrenia and
comorbid substance use disorders (Barrowclough et al., 2001).
Definition
CBT was defined as a discrete psychological intervention where service users:
• establish links between their thoughts, feelings or actions with respect
to the current or past symptoms, and/or functioning, and
• re-evaluate their perceptions, beliefs or reasoning in relation to the
target symptoms.
In addition, a further component of the intervention should involve the
following:
• service users monitoring their own thoughts, feelings or behaviours
with respect to the symptom or recurrence of symptoms, and/or
• promotion of alternative ways of coping with the target symptom,
and/or
• reduction of distress, and/or
• improvement of functioning.

eligibility criteria, can be found in Table 65. The primary clinical questions can be
found in Box 1. For the 2009 guideline, a new systematic search was conducted for
relevant RCTs published since the 2002 guideline (further information about the
search strategy can be found in Appendix 20 and information about the search for
health economic evidence can be found in Section 9.4.8).

In the 2002 guideline, 13 RCTs (N = 1,297) of CBT were included. One RCT from the
2002 guideline (KEMP1996) was removed from the 2009 guideline analysis and re-
classified by the GDG as adherence therapy and a further three studies were
removed because of inadequate numbers of participants (Garety1994; Levine1996;
Turkington2000). The search for the 2009 guideline identified six papers providing
follow-up data to existing RCTs and 22 new RCTs, including those with CBT as part
of a multi-modal intervention. In total, 31 RCTs (N = 3,052) met the inclusion criteria
for the update. Of these, one was currently unpublished and 30 were published in
peer-reviewed journals between 1996 and 2008 (further information about both
included and excluded studies can be found in Appendix 22c).
Table 65: Clinical review protocol for the review of CBT

Study design RCT (≥ 10 participants per arm)
Patient population Adults (18+) with schizophrenia (including
schizophrenia-related disorders)
Excluded populations Very late onset schizophrenia (onset after age60) Other
psychotic disorders, such as bipolar disorder, mania or
depressive psychosis
People with coexisting learning difficulties, significant
physical or sensory difficulties, or substance misuse
Interventions CBT
Adherence to antipsychotic treatment
Insight
Quality of life
Adverse events
9.4.4 Cognitive behavioural therapy versus control
For the 2009 guideline review, 31 RCTs of CBT versus any type of control were
included in the meta-analysis (see Table 66 for a summary of the study
characteristics). However this comparison was only used for outcomes in which
there were insufficient studies to allow for a separate standard care and other active
treatment arms.

For the primary analysis, 19 RCTs were included comparing CBT with standard care,
14 comparing CBT with other active treatments and three comparing CBT with non-
standard care. Forest plots and/or data tables for each outcome can be found in
Appendix 23d.
In addition to the primary analyses, subgroup analyses were used to explore certain
characteristics of the trials 24 (see Table 67 for a summary of the studies included in
each subgroup comparison). Five RCTs were included in the analysis comparing
CBT with any control in participants experiencing a first episode of schizophrenia;
eight compared CBT with any control in participants experiencing an acute-episode;
11 compared CBT with any control in participants during the promoting recovery
phase; six compared group CBT with any control; and 19 compared individual CBT
with any control. Multi-modal trials were not included in the subgroup analyses.
Forest plots and/or data tables for each outcome can be found in Appendix 23d.
24Existing subgroup comparisons assessing the country of the trial, number of treatment sessions
and duration of treatment were also updated. However, there was insufficient data to draw any
conclusions based on these subgroups. Please refer to Appendix 23d for the forest plots and/or data
tables for all subgroup comparisons conducted.

Table 66: Summary of study characteristics for CBT
CBT versus any controla CBT versus standard care CBT versus other active CBT versus
treatments non-standard care
k (total N) 31 (3052) 19 (2118) 14 (1029) 3 (136)

Study ID BACH2002 BACH2002 BECHDOLF2004 Drury1996
BARROW-CLOUGH2006 BARROW-CLOUGH2006 CATHER2005 Bradshaw2000
BECHDOLF2004 DURHAM2003 DURHAM2003 RECTOR2003
Bradshaw2000 ENGLAND2007 GARETY2008
CATHER2005 GARETY2008 Haddock1999
Drury1996 GRANHOLM2005c Hogarty1997
DURHAM2003 GUMLEY2003 JACKSON2007
ENGLAND2007 JACKSON2005 LECOMTE2008
GARETY2008b JENNER2004c Lewis2002
GRANHOLM2005c Kuipers1997 PENADES2006
GUMLEY2003 LECLERC2000 PINTO1999c
Haddock1999 LECOMTE2008 Sensky2000
Hogarty1997e Lewis2002 Tarrier1998
JACKSON2005 MCLEOD2007 VALMAGGIA2005
JACKSON2007 STARTUP2004
JENNER2004c Tarrier1998
Kuipers1997 TROWER2004
LECLERC2000 Turkington2002
LECOMTE2008 WYKES2005
Lewis2002d
MCLEOD2007

PENADES2006
PINTO1999c
RECTOR2003
Sensky2000
STARTUP2004
Tarrier1998
TROWER2004
Turkington2002
VALMAGGIA2005
WYKES2005
Diagnosis 58–100% 58–100% 64–100% 100% schizophrenia or other
Schizophrenia or Schizophrenia or Schizophrenia or related diagnoses (DSM or
other related diagnoses Other related diagnoses Other related diagnoses ICD-10)
(DSM or ICD-10) (DSM or ICD-10) (DSM or ICD-10)
Baseline severity Not reported
BPRS total: BPRS total: PANSS total:
Mean (SD) range: Mean (SD) range: Mean (SD) range:
~17 (7) to ~82 (21) ~17 (7) to ~82 (21) ~51 (13) to~96 (16)
PANSS total: PANSS total: CPRS total:

Mean (SD) range: Mean (SD) range: Mean (SD) ~36 (14)
~25 (7) to ~96 (16) ~25 (7) to ~96 (16)
CPRS total: CPRS total:

Mean (SD) ~24 (14) to ~36 (14) Mean (SD) range: ~24 (14)
Continued

Number of sessions Range: 4–156 Range: 4–24 Range: 10–156 Range: 20–156
Length of treatment Range: 2–156 weeks Range: 2–52 weeks Range: 8–156 weeks Range: 24–156 weeks
Length of follow-up Range: 3–60 months Range: 3–60 months Range: 3–60 months Range: 6–24 months
(only including
papers reporting
follow-up measures)
Setting Inpatient: Inpatient: Inpatient: Inpatient:

BECHDOLF2004 Lewis2002f BECHDOLF2004 Bradshaw2000
Bradshaw2000 STARTUP2004 Haddock1999 Drury1996
Drury1996 Hogarty1997e Lewis2002f
Haddock1999 VALMAGGIA2005
Hogarty1997e
Lewis2002f
STARTUP2004
VALMAGGIA2005
Outpatient: Outpatient: Outpatient:
Outpatient:
BARROW-CLOUGH2006 CATHER2005 RECTOR2003
BARROW-CLOUGH2006
ENGLAND2007 LECOMTE2008
CATHER2005
GRANHOLM2005c Sensky2000
ENGLAND2007
GUMLEY2003 Tarrier1998
GRANHOLM2005c
JACKSON2005
GUMLEY2003

Table 66: (Continued)
k (total N) 31 (3052) 19 (2118) 14 (1029) 3 (136)

Study ID BACH2002 BACH2002 BECHDOLF2004 Drury1996
BARROW-CLOUGH2006 BARROW-CLOUGH2006 CATHER2005 Bradshaw2000
BECHDOLF2004 DURHAM2003 DURHAM2003 RECTOR2003
Bradshaw2000 ENGLAND2007 GARETY2008
CATHER2005 GARETY2008 Haddock1999
Drury1996 GRANHOLM2005c Hogarty1997
DURHAM2003 GUMLEY2003 JACKSON2007
ENGLAND2007 JACKSON2005 LECOMTE2008
GARETY2008b JENNER2004c Lewis2002
GRANHOLM2005c Kuipers1997 PENADES2006
GUMLEY2003 LECLERC2000 PINTO1999c
Haddock1999 LECOMTE2008 Sensky2000
Hogarty1997e Lewis2002 Tarrier1998
JACKSON2005 MCLEOD2007 VALMAGGIA2005
JACKSON2007 STARTUP2004
JENNER2004c Tarrier1998
Kuipers1997 TROWER2004
LECLERC2000 Turkington2002
LECOMTE2008 WYKES2005
Lewis2002d
MCLEOD2007

JACKSON2005 JENNER2004c
JENNER2004c Kuipers1997
Kuipers1997 LECOMTE2008
LECOMTE2008 Sensky2000
RECTOR2003 Tarrier1998
Sensky2000 WYKES2005
Tarrier1998
WYKES2005
Inpatient and outpatient: Inpatient and outpatient: Inpatient and outpatient:

BACH2002 BACH2002 DURHAM2003
DURHAM2003 DURHAM2003 GARETY2008
GARETY2008 GARETY2008 PINTO1999c
LECLERC2000 LECLERC2000
MCLEOD2007 MCLEOD2007
PINTO1999c TROWER2004
TROWER2004 Turkington2002
Turkington2002 EIS setting:
EIS setting: JACKSON2007
JACKSON2007
Note. Studies were categorised as short (fewer than 12 weeks), medium (12–51 weeks) and long (52 weeks or more).
aCBT versus any control was only used for outcomes in which there were insufficient studies to allow for separate standard care and other
active treatment arms.

bThe primary GARETY2008 paper reports data separately for the carer and non-carer pathways of the study. Although the dichotomous data
has been combined across pathways, data for the continuous measures are presented separately. In the main and subgroup analyses
GARETY2008 appears as GARETY2008C (carer pathway) and GARETY2008NC (non-carer pathway).
cMulti-modal interventions.
dFollow-up papers to Lewis2002 report the data separately for the three study sites, hence in the analysis Lewis 2002 appears as LEWIS2002L
(Liverpool), LEWIS2002M (Manchester) and LEWIS2002N (Nottingham).

eParticipants were recruited in the inpatient setting with the intervention starting shortly before discharge.
fParticipants were recruited from inpatient wards and day hospitals.

9.4.5 Training
The inconsistency in reporting what training the therapists in the trials had received
meant it was impossible to determine the impact of level of training on the outcomes
of the trial. Less than half (15/31) of the included CBT papers made reference to
specific CBT-related training. In early CBTp trials this is not surprising because the
researchers were at the forefront of the development of the therapy and no specific
psychosis-related CBT training would have been available. In studies where training
was mentioned, it was often vague in terms of the length of training therapists had
received and whether the training had been specifically focused on CBT for
psychosis. Moreover, where details of training programmes associated with the trial
were provided, previous experience and training did not always appear to have been
controlled for. This means that therapists could have entered the study with different
levels of competence, making it impossible to determine the impact of the specified
training programme. Of the 25 trials reporting the professional conducting the
intervention, the majority utilised clinical psychologists (14/25). However, a
proportion of trials utilised different professionals including psychiatrists (3/25),
psychiatric nurses (7/25), social workers (2/25), Master’s level psychology graduates
and/or interns (1/25), occupational therapists (1/24) and local mental health
workers (2/25). Within some trials, a number of professionals may have delivered
the intervention (for example, two psychologists and one psychiatrist). Often, where
the professional conducting the intervention was not a clinical psychologist,
reference was made to specific training in CBTp or extensive experience working
with people with psychosis.

Table 67: Summary of study characteristics for CBT subgroup analyses
CBT versus any CBT versus any CBT versus any Group CBT versus Individual CBT
control– first episodea control– acute episode control– promoting any control versus any control
recovery
K (total N) 5 (618) 8 (695) 11 (1093) 6 (534) 19 (2082)

Study ID Haddock1999 BACH2002 BARROW- BARROW- BACH2002
JACKSON2005 BECHDOLF2004 CLOUGH2006 CLOUGH2006 Bradshaw1999
JACKSON2007 Bradshaw2000 CATHER2005 BECHDOLF2004 CATHER2005
LECOMTE2008 Drury1996 DURHAM2003 LECOMTE2008 DURHAM2003
Lewis2002 ENGLAND2007 Kuipers1997 LECLERC2000 ENGLAND2007
GARETY2008 PENADES2006 MCLEDO2007 GARETY2008
MCLEOD2007 Sensky2000 WYKES2005 GUMLEY2003
STARTUP2004 Tarrier1998 Haddock1999
TROWER2004 JACKSON2005
Turkington2002 JACKSON2007
VALMAGGIA2005 Kuipers1997
WYKES2005 Lewis2002
PENADES2006
Sensky2000
STARTUP2004
Tarrier1998
TROWER2004
Turkington2002
VALMAGGIA2005
Note. Studies were categorised as short (<12 weeks), medium (12–51 weeks) and long (52 weeks or more).
aA number of trials included participants in all phases of illness (for example, 20% first episode, 60% acute and 20% promoting recovery)
and hence could not be included in the subgroup analysis.

Competence does not appear to be directly correlated with training and a number of
additional variables play a part. The Durham and colleagues’(2003) study indicated
that training in general CBT did not necessarily produce proficient CBTp therapists.
Although the therapists in the study had undergone CBT training, when their
practice was assessed on a CBTp fidelity measure, they did not appear to be using
specific psychosis-focused interventions. A number of studies included in the CBTp
meta-analyses used CBT fidelity measures to determine the quality of the therapy
that was being delivered. Again, there were inconsistencies between studies. Three
different fidelity measures were used and there was no agreed standard as to what
the cut- off score for demonstrating competence should be. Moreover, Durham and
colleagues (2003) used two of these scales in their trial and found that therapy
ratings did not correlate.
With regard to the use of treatment manuals, however, there was more consistent
reporting across the trials, with the majority of papers (24/31) making reference to
either a specific treatment manual or to a manualised approach. Reporting of
supervision was also more consistent, with both peer- and senior-supervision
evident in over two-thirds of the trials.
9.4.6 Ethnicity
Only one follow-up paper (Rathod et al., 2005) assessed changes in insight and
compliance in the Black Caribbean and African–Caribbean participants included in
the Turkington 2002 study. The subgroup analysis indicated a higher dropout rate
among both black and ethnic minority groups. Additionally, compared with their
white counterparts, the black and minority ethnic participants demonstrated
significantly smaller changes in insight. Although these are potentially interesting
findings, it must be noted that black and minority ethnic participants comprised only
11% of the study population, with Black African and African–Caribbean participants
representing 3 and 5% of the sample, respectively. With regard to the other studies
included in the review, there was a paucity of information on the ethnicity of
participants. Because of the lack of information, the GDG were unable to draw any
conclusions from the data or make any recommendations relating to practice.
However, the GDG acknowledge that this is an area warranting further research and
formal investigation.

The review found consistent evidence that, when compared with standard care, CBT
was effective in reducing rehospitalisation rates up to 18 months following the end
of treatment. Additionally, there was robust evidence indicating that the duration of
hospitalisation was also reduced (8.26 days on average). Consistent with the 2002
guideline, CBT was shown to be effective in reducing symptom severity as measured
by total scores on items, such as the PANSS and BPRS, both at end of treatment and
at up to 12 months’ follow-up. Robust small to medium effects (SMD~0.30) were also
demonstrated for reductions in depression when comparing CBT with both standard
care and other active treatments. Furthermore, when compared with any control,
there was some evidence for improvements in social functioning up to 12 months.

Although the evidence for positive symptoms was more limited, analysis of
PSYRATS data demonstrated some effect for total hallucination measures at the end
of treatment. Further to this, there was some limited but consistent evidence for
symptom-specific measures including voice compliance, frequency of voices and
believability, all of which demonstrated large effect sizes at both end of treatment
and follow-up. However, despite these positive effects for hallucination-specific
measures, the evidence for there being any effect on delusions was inconsistent.
Although no RCTs directly compared group-based with individual CBT, indirect
comparisons indicated that only the latter had robust effects on rehospitalisation,
symptom severity and depression. Subgroup analyses also demonstrated additional
effects for people with schizophrenia in the promoting recovery phase both with and
without persistent symptoms. In particular, when compared with any other control,
studies recruiting people in the promoting recovery phase demonstrated consistent
evidence for a reduction in negative symptoms up to 24 months following the end of
treatment.

The systematic literature search identified two economic studies that assessed the
cost effectiveness of CBT for people with schizophrenia (Kuipers et al., 1998; Startup
et al., 2005). Both studies were undertaken in the UK. Details on the methods used
for the systematic search of the economic literature are described in Appendix 11.
References to included/excluded studies and evidence tables for all economic
studies included in the guideline systematic literature review are presented in the
form of evidence tables in Appendix 25.
Kuipers and colleagues (1998) evaluated the cost effectiveness of CBT added to
standard care compared with standard care alone in 60 people with medication-
resistant psychosis participating in an RCT conducted in the UK (KUIPERS1997).
The time horizon of the analysis was 18 months (RCT period plus naturalistic follow-
up). The study estimated NHS costs (inpatient, outpatient, day hospital, primary and
community services) and costs associated with specialist, non-domestic
accommodation. Medication costs were not considered. The primary outcome of the
analysis was the mean change in BPRS score. CBT was shown to be significantly
more effective than its comparator in this respect, with the treatment effect lasting 18
months after the start of the trial (p <0.001). The costs between the two treatment
groups were similar: the mean monthly cost per person over 18 months was £1,220
for CBT added to standard care and £1,403 for standard care alone (p =0.416, 1996
prices). The study had in sufficient power to detect significant differences in costs.
The authors suggested that CBT might be a cost-effective intervention in medication-
resistant psychosis, as the clinical benefits gained during the 9 months of CBT were
maintained and even augmented 9 months later, while the extra intervention costs
seemed to be offset by reduced utilisation of health and social care services.

Startup and colleagues (2005) conducted a cost-consequence analysis to measure the
cost effectiveness of CBT on top of treatment as usual versus treatment as usual
alone in 90 people hospitalised for an acute psychotic episode participating in an
RCT in North Wales (STARTUP2004). The time horizon of the analysis was 2 years;
the perspective was that of the NHS and Personal Social Services (PSS). Costs
included hospital, primary, community and residential care and medication. Health
outcomes were measured using the Scale for the Assessment of Positive Symptoms
(SAPS), the Scale for the Assessment of Negative Symptoms (SANS), the Social
Functioning Scale (SFS) and the GAF scale. CBT showed a significant effect over
control in SANS and SFS scores, at no additional cost: the mean cost per person over
24 months was £27,535 for the CBT group and £27,956 for the control group (p =
0.94). The study had insufficient power for economic analysis.
The above results indicate that CBT is potentially a cost-effective intervention for
people with acute psychosis or medication-resistant schizophrenia. However, the
study samples were very small in both studies and insufficient to establish such a
hypothesis with certainty.
Economic modelling
Objective
The guideline systematic review and meta-analysis of clinical evidence
demonstrated that provision of CBT to people with schizophrenia results in clinical
benefits and reduces the rates of future hospitalisation. A cost analysis was
undertaken to assess whether the costs to the NHS of providing CBT in addition to
standard care to people with schizophrenia are offset by future savings resulting
from reduction in hospitalisation costs incurred by this population.
Intervention assessed
According to the guideline systematic review and meta-analysis of clinical evidence,
group-based CBT is not an effective intervention. Therefore, the economic analysis
compared individually-delivered CBT added to standard care versus standard care
alone.
Methods
A simple economic model estimated the net total costs (or cost savings) to the NHS
associated with provision of individual CBT in addition to standard care to people
with schizophrenia. Two categories of costs were assessed: intervention costs of CBT,
and cost savings resulting from the expected reduction in hospitalisation rates in
people with schizophrenia receiving CBT, estimated based on the guideline meta-
analysis of respective clinical data. Standard care costs were not estimated, because
these were common to both arms of the analysis.
Cost data
Intervention costs (costs of providing CBT)
The clinical studies on individual CBT included in the guideline systematic review
described programmes of varying numbers of sessions. The resource use estimate

associated with provision of CBT in the economic analysis was based on the average
resource use reported in these studies, confirmed by the GDG expert opinion to be
consistent with clinical practice in the UK. According to the reported resource use data,
CBT in the economic analysis consisted of 16 individually-delivered sessions lasting 60
minutes each.
CBT can be delivered by a variety of mental health professionals with appropriate

training and supervision. The salary level of a mental health professional providing
CBT was estimated by the GDG to range between bands 6 and band8. This is
comparable with the salary level of a clinical psychologist. Therefore, the unit cost of
clinical psychologists was used to estimate an average intervention cost. The unit
cost of a clinical psychologist has been estimated at £67 per hour of client contact in
2006/07prices (Curtis, 2007). This estimate has been based on the mid-point of
Agenda for Change salary band7 of the April 2006 payscale according to the
National Profile for Clinical Psychologists, Counsellors and Psychotherapists (NHS
Employers, 2006). It includes salary, salary oncosts, overheads and capital overheads
but does not take into account qualification costs because the latter are not available
for clinical psychologists. The same source of national health and social care unit
costs reports the cost of CBT as £67 per hour of face-to-face contact ((Curtis, 2007);
2006/07price). This latter unit cost has been estimated on the basis that CBT is
delivered by a variety of health professionals, including specialist registrars, clinical
psychologists and mental health nurses, and is equal to the unit cost of a clinical
psychologist per hour of client contact.
Based on the above resource use estimates and the unit cost of clinical psychologists,
the cost of providing a full course of CBT to a person with schizophrenia was
estimated at £1,072 in 2006/07prices.
Costs of hospitalisation / cost savings from reduction in hospitalisation rates The average
cost of hospitalisation for a person with schizophrenia was estimated by multiplying
the average duration of hospitalisation for people with schizophrenia, schizotypal
and delusional disorders in England in 2006/07 (NHS The Information Centre,
2008b) by the national average unit cost per bed-day in an inpatient mental health
acute care unit for adults for 2006/07 (NHS Reference Costs, (Department of Health,
2008)). Hospital Episode Statistics (HES) is a service providing national statistical
data of the care provided by NHS hospitals and for NHS hospital patients treated
elsewhere in England (NHS The Information Centre, 2008b). With respect to
inpatient data, HES records episodes (periods) of continuous admitted patient care
under the same consultant. In cases where responsibility for a patient’s care is
transferred to a second or subsequent consultant, there will be two or more episodes
recorded relating to the patient’s stay in hospital. This means that, for any condition
leading to hospital admission, the average length of inpatient stay as measured and
reported by HES may be an underestimation of the actual average duration of
continuous hospitalisation. Based on HES, the average duration of hospitalisation for
people with schizophrenia, schizotypal and delusional disorders (F20–F29 according
to ICD-10) in England was 110.6 days in 2006/07. Based on the annually collected

NHS Reference Costs (NHS The Information Centre, 2008b) the cost per bed-day
in a mental health acute care inpatient unit was £259 in 2006/07. By multiplying
these figures, the average cost of hospitalisation per person with schizophrenia was
estimated at £28,645 in 2006/07prices.
Clinical data on hospitalisation rates following provision of cognitive behavioural therapy

The guideline meta-analysis of CBT data on hospitalisation rates showed that
providing CBT in addition to standard care to people with schizophrenia
significantly reduces the rate of future hospitalisations compared with people
receiving standard care alone. Table 68 shows the CBT studies included in the meta-
analysis of hospitalisation-rate data up to 18 months following treatment (whether
these studies were conducted in the UK or not), the hospitalisation rates for each
treatment arm reported in the individual studies and the results of the meta-analysis.
The results of meta-analysis show that CBT, when added to standard care, reduces
the rate of future hospitalisations in people with schizophrenia (RR of hospitalisation
of CBT added to standard care versus standard care alone: 0.74). This result was
statistically significant at the 0.05 level ( 95% CIs of RR: 0.61 to 0.94).
The baseline rate of hospitalisation in the economic analysis was taken from the
overall rate of hospitalisation under standard care alone as estimated in the
guideline meta-analysis of CBT data on hospitalisation rates; that is, a 29.98%
baseline hospitalisation rate was used. The rate of hospitalisation when CBT was
added to standard care was calculated by multiplying the estimated RR of
hospitalisation of CBT plus standard care versus standard care alone by the baseline
hospitalisation rate.
Details on the clinical studies considered in the economic analysis are available in
Appendix 22c. The forest plots of the respective meta-analysis are provided in
Appendix 23d.
Table 68: Studies considered in the economic analysis of CBT in addition to

standard care versus standard care alone and results of meta-analysis
Study ID Country Total events (n) in each treatment arm (N)

CBT plus standard care Standard care alone (n/N)
(n/N)
TARRIER1998 UK 16/33 9/28
BACH2002 Non-UK 12/40 19/40
LEWIS2002 UK 33/101 37/102
TURKINGTON2002 UK 36/257 38/165
GUMLEY2003 UK 11/72 19/72
Total 108/503 (21.47%) 122/407 (29.98%)
Meta-analysis results RR: 0.74
95% CI: 0.61–0.94

Sensitivity analysis
One-way sensitivity analyses were undertaken to investigate the robustness of the
results under the uncertainty characterising some of the input parameters and the
use of different data and assumptions in the estimation of total net costs (or net
savings) associated with provision of CBT to people with schizophrenia. The
following scenarios were explored:
• use of the 95% CIs of the RR of hospitalisation of CBT added to
standard care versus standard care alone
• exclusion of TARRIER1998 from the meta-analysis. TARRIER1998 was
carried out before the National Service Framework was implemented,
and therefore the way the study was conducted in terms of
hospitalisation levels may have been different from current clinical
practice. The baseline rate of hospitalisation used in the analysis was
the pooled, weighted, average hospitalisation rate of the control arms
of the remaining studies
• exclusion of BACH2002 from the meta-analysis as this was a non-UK
study and clinical practice regarding hospital admission levels may
have been different from that in the UK. The baseline rate of
hospitalisation used in the analysis was the pooled, weighted, average
hospitalisation rate of the control arms of the remaining studies
• exclusion of both TARRIER1998 and BACH2002 from the meta-
analysis. The baseline rate of hospitalisation used in the analysis was
the pooled, weighted, average hospitalisation rate of the control arms
of the remaining studies
• change in the number of CBT sessions (16 in the base-case analysis) to a
range between 12 and 20
• change in the baseline rate of hospitalisation (that is, the hospitalisation
rate for standard care which was 29.98% in the base-case analysis) to a
range between 20 and 40%
• use of a more conservative value of duration of hospitalisation. The
average duration of hospitalisation for people with schizophrenia (ICD
F20-F29) reported by HES (NHS The Information Centre, 2008b) was
110.6 days, which was deemed high by the GDG. Indeed, HES reported
a median duration of hospitalisation for this population of 36 days.
HES data were highly skewed, apparently from a number of people
with particularly long hospital stays. An alternative, lower length of
hospitalisation of 69 days was tested, taken from an effectiveness trial
of clozapine versus SGAs in people with schizophrenia with
inadequate response or intolerance to current antipsychotic treatment
conducted in the UK (CUtLASS Band 2, (Davies et al., 2008)).
Results
Base-case analysis
The reduction in the rates of future hospitalisation achieved by offering CBT to
people with schizophrenia in addition to standard care yielded cost savings

equalling £2,061 per person. Given that provision of CBT costs £1,072 per person,
CBT results in an overall net saving of £989 per person with schizophrenia. Full
results of the base-case analysis are reported in Table 69.
Table 69: Results of cost analysis comparing CBT in addition to standard care
versus standard care alone per person with schizophrenia
Costs CBT plus standard Standard care alone Difference

care
CBT cost £1,072 0 £1,072
Hospitalisation cost £6,526 £8,587 −£2,061
Total cost £7,598 £8,587 −£989
The results of the base-case analysis were overall robust to the different scenarios
explored in sensitivity analysis. When the 95% CIs of the RR of hospitalisation were
used, then the total net cost of providing CBT ranged from −£2,277 (that is a net
saving) to £557 per person. When the more conservative value of 69 days length of
hospitalisation (instead of 110.6 days used in the base-case analysis) was tested, the
net cost of providing CBT ranged between −£1,017 (net saving) to £751 per person. In
all scenarios, using the relevant mean RR of hospitalisation taken from the guideline
meta-analysis, addition of CBT to standard care resulted in overall cost savings
because of a substantial reduction in hospitalisation costs. It must be noted that
when BACH2002 was excluded from analysis, then the results of meta-analysis were
insignificant at the 0.05 level; consequently, when the upper 95% CI of RR of
hospitalisation was used, CBT added to standard care incurred higher
hospitalisation costs relative to standard care alone.
Full results of sensitivity analysis are presented in Table 70.
Discussion
The economic analysis showed that CBT is likely to be an overall cost-saving
intervention for people with schizophrenia because the intervention costs are offset
by savings resulting from a reduction in the number of future hospitalisations
associated with this therapy. The net cost of providing CBT was found to lie between
−£2,277 (overall net saving) and £557 per person with schizophrenia (for a mean
duration of hospitalisation of 110.6 days) or −£1,017 to £751 per person (for a mean
duration of hospitalisation of 69 days), using the 95% CIs of RRs of hospitalisation,
as estimated in the guideline meta-analysis. It must be noted that possible reduction
in other types of health and social care resource use and subsequent cost savings to
the NHS and social services, as well as broader financial implications to society (for
example, potential increased productivity) associated with the provision of CBT to
people with schizophrenia, have not been estimated in this analysis. In addition,
clinical benefits associated with CBT, affecting both people with schizophrenia and
their families/carers, such as symptom improvement and enhanced HRQoL
following reduction in future inpatient stays, should also be considered when the

cost effectiveness of CBT is assessed. Taking into account such benefits, even a
(conservative) net cost of £751 per person can be probably justified.
Table 70: Results of sensitivity analysis of offering CBT in addition to standard

care to people with schizophrenia
Scenario Total net cost (negative cost implies net

saving)
Use of 95% CIs of RR of hospitalisation −£2,277 (lower CI) to £557 (upper CI)
Exclusion of TARRIER1998 from meta- −£1,490 (−£2,771 to £47 using the

analysis
95% CIs of RR of hospitalisation)
Exclusion of BACH2002 −£375 (−£2,465 to £2,599 using the
(non-UK study) from meta-analysis
Exclusion of TARRIER1998 and −£1,231 (−£2,502 to £437 using the
BACH2002 from meta-analysis
CBT sessions between 12 and 20 −£1,257 to −£721, respectively
Hospitalisation rate understandard care −£1,678 to−£303, respectively
between 40 and 20%
Mean length of hospitalisation −£214 (−£1,017 to £751 using the 95%
69 days
CIs of RR of hospitalisation)

The conclusions drawn in the 2002 guideline regarding the efficacy of CBT have
been supported by the 2009 systematic review. The data for the reduction in
rehospitalisation rates and duration of admission remains significant even when
removing non-UK and pre-National Service Framework for Mental Health
(Department of Health, 1999) papers in a sensitivity analysis, suggesting that these
findings may be particularly robust within the current clinical context. The
effectiveness of CBT has been corroborated by the evidence for symptom severity,
which included reductions in hallucination-specific measures and depression in
addition to total symptom scores. However, it must be noted that despite general
confirmation of the 2002 recommendations, following the reclassification and
subsequent removal of KEMP1996, there was no robust evidence for the efficacy of
CBT on measures of compliance or insight. Consequently, the GDG concluded that
there is insufficient evidence to support the 2002 recommendation about the use of
CBT to assist in the development of insight or in the management of poor treatment
adherence.
The systematic review of economic evidence showed that provision of CBT to people
with schizophrenia in the UK improved clinical outcomes at no additional cost. This
finding was supported by economic modelling undertaken for this guideline, which
suggested that provision of CBT might result in net cost savings to the NHS,
associated with a reduction in future hospitalisation rates. The results of both the
systematic literature review and the economic modelling indicate that providing

individual CBT to people with schizophrenia is likely to be cost effective in the UK
setting, especially when clinical benefits associated with CBT are taken into account.
Although the GDG were unable to draw any firm conclusions from subgroup
analyses assessing the impact of treatment duration and number of sessions, they
did note that the evidence for CBT is primarily driven by studies that included at
least 16 planned sessions. To incorporate the current state of evidence and expert
consensus, the GDG therefore modified the 2002 recommendation relating to the
duration and number of treatment sessions.
There was, however, more reliable evidence to support the provision of CBT as an
individual-based therapy, a finding largely consistent with current therapeutic
practice within the UK.
From the CBTp studies included in the meta-analyses, it is not possible to make any
recommendations on the specific training requirements or competencies required to
deliver effective CBTp. In particular, papers varied widely in the degree to which
they reported details about the training and experience of the person delivering the
intervention. However, the GDG felt that this is an important area for future
development and have made a research recommendation. Despite not being able to
make any specific recommendations for the types of training required at this stage, it
was noted that, overall, the majority of trials used either clinical psychologists or
registered and/or accredited psychological therapists to deliver the CBTp. In
addition, regular clinical supervision was provided in two thirds of the trials and
treatment manuals utilised in nearly all of the trials. From this evidence, and based
upon expert opinion, the GDG included a number of recommendations relating to
the delivery of CBT for people with schizophrenia.
Both the consistency with which CBT was shown to be effective across multiple
critical outcomes and the potential net cost-savings to the NHS support the 2002
recommendations regarding the provision of CBT to people with schizophrenia.
**2009**
For the 2014 guideline the GDG took the view that, following the publication of
Psychosis and Schizophrenia in Children and Young People, the 2014 guideline should be
consistent where appropriate, including changing the population from ‘people with
schizophrenia’ to ‘people with psychosis and schizophrenia’. Therefore the GDG
saw the value in advising practitioners of the equivocal evidence regarding
psychological interventions when compared with antipsychotic medication and
recommended that if a person wished to try a psychological intervention alone, this
could be trialled over the course of 1 month or less. The GDG also wished to make it
explicit that the options for first episode psychosis and for an acute exacerbation or
recurrence of psychosis or schizophrenia should be psychological interventions
(individual CBT and family intervention) combined with oral antipsychotic
medication.

Treatment options for first episode psychosis
9.4.10.1 For people with first episode psychosis offer:
• oral antipsychotic medication (see recommendations 10.11.1.2–10.11.1.13) in
conjunction with
• psychological interventions (family intervention and individual CBT,
delivered as described in recommendations 9.4.10.3 and 9.7.10.3). [new 2014]
9.4.10.2 Advise people who want to try psychological interventions alone that these
are more effective when delivered in conjunction with antipsychotic
medication. If the person still wants to try psychological interventions alone:
• offer family intervention and CBT
• agree a time (1 month or less) to review treatment options,
including introducing antipsychotic medication
• continue to monitor symptoms, distress, impairment and level of
functioning (including education, training and employment)
regularly. [new 2014]
How to deliver psychological interventions

9.4.10.3 CBT should be delivered on a one-to-one basis over at least 16 planned
sessions and:
• follow a treatment manual 25 so that:
- people can establish links between their thoughts, feelings or
actions and their current or past symptoms, and/or functioning
- the re-evaluation of people’s perceptions, beliefs or reasoning
relates to the target symptoms
• also include at least one of the following components:
- people monitoring their own thoughts, feelings or behaviours with
respect to their symptoms or recurrence of symptoms
- promoting alternative ways of coping with the target symptom
- reducing distress
- improving functioning. [2009]

9.4.10.4 For people with an acute exacerbation or recurrence of psychosis or
schizophrenia, offer:
conjunction with
25 Treatment manuals that have evidence for their efficacy from clinical trials are preferred.

9.4.10.5 Offer CBT to all people with psychosis or schizophrenia (delivered as
described in recommendation 9.4.10.3). This can be started either during the
acute phase or later, including in inpatient settings. [2009]
Promoting recovery
9.4.10.6 Offer CBT to assist in promoting recovery in people with persisting positive
and negative symptoms and for people in remission. Deliver CBT as
described in recommendation 9.4.10.3. [2009]

9.4.11.1 An adequately powered RCT should be conducted to investigate the most
appropriate duration and number of sessions for CBT in people with
schizophrenia.[2009]
9.4.11.2 An adequately powered RCT should be conducted to investigate CBT
delivered by highly trained therapists and mental health professionals
compared with brief training of therapists in people with
9.4.11.3 Research is needed to identify the competencies required to deliver effective
CBT to people with schizophrenia.[2009]
9.5 COGNITIVE REMEDIATION

9.5.1 Introduction
**2009** The presence of cognitive impairment in a proportion of people with
schizophrenia has been recognised since the term ‘schizophrenia’ was first coined
(Bleuler, 1911). The precise cause of these deficits (such as structural brain changes,
disruptions in neuro-chemical functions or the cognitive impact of the illness and/or
of medication) remains contentious, whereas progress on characterising the
cognitive problems that arise in schizophrenia has been substantial. Major domains
identified include memory problems (Brenner, 1986), attention deficits (Oltmanns &
Neale, 1975) and problems in executive function, such as organisation and planning
(Weinberger et al., 1988). A recent initiative to promote standardisation of methods
for evaluating research on cognitive outcomes (the Measurement and Treatment
Research to Improve Cognition in Schizophrenia consensus panel [MATRICS;
(Nuechterlein et al., 2004)]) has identified eight more specific domains:
attention/vigilance; speed of processing; working memory; verbal learning and
memory; visual learning and memory; reasoning and problem solving; verbal
comprehension; and social cognition. Few studies as yet examine changes in all these
domains. Cognitive impairment is strongly related to functioning in areas such as
work, social relationships and independent living (McGurk et al., 2007). Because of
the importance of cognitive impairment in terms of functioning, it has been
identified as an appropriate target for interventions.
Currently available pharmacological treatments have limited effects on cognitive

impairments (see Chapter 10). Cognitive remediation programmes have therefore

been developed over the past 40 years with the goal of testing whether direct
attempts to improve cognitive performance might be more effective (McGurk et al.,
2007). The primary rationale for cognitive remediation is to improve cognitive
functioning, with some papers also stating improved functioning as an additional
aim (Wykes & Reeder, 2005). Approaches adopted have ranged from narrowly
defined interventions, which involve teaching service users to improve their
performance on a single neuropsychological test, to the provision of comprehensive
remediation programmes, increasingly using computerised learning (Galletly et al.,
2000). The programmes employ a variety of methods, such as drill and practice
exercises, teaching strategies to improve cognition, suggesting compensatory
strategies to reduce the effects of persistent impairments and group discussions
(McGurk et al., 2007).
Because the use of these methods in the treatment of schizophrenia is still

developing and early studies had mixed results (Pilling et al., 2002), there remains
uncertainty over which techniques should be used (Wykes & van der Gaag, 2001)
and whether the outcomes are beneficial, both in terms of sustained effects on
cognition and for improving functioning. Reports of combinations of cognitive
remediation with other psychosocial interventions, such as social skills training, or
vocational interventions, such as supported employment programmes, have been
increasing in the literature. In this review, the focus is on cognitive remediation as a
single-modality intervention except where it has been combined with another of the
psychological or psychosocial interventions. In these cases, the intervention has been
classified as multi-modal intervention and subjected to sensitivity analyses (see
Section 9.1.5).**2009** A review of cognitive remediation combined with any
vocational rehabilitation interventions can be found in Chapter 13.
Definition
**2009**Cognitive remediation was defined as:
• an identified procedure that is specifically focused on basic cognitive
processes, such as attention, working memory or executive
functioning, and
• having the specific intention of bringing about an improvement in the
level of performance on that specified cognitive function or other
functions, including daily living, social or vocational skills.

eligibility criteria can be found in
Table 71. The primary clinical questions can be found in Box 1. For the 2009
guideline, a new systematic search was conducted for relevant RCTs published since
the 2002 guideline (further information about the search strategy can be found in
Appendix 20). It must be acknowledged that some cognitive remediation studies cite
improvements to cognition/cognitive measures as their primary outcome. However,
it is the view of the GDG that only sustained improvements in cognition, as

measured at follow-up, should be considered as clinically important. The rationale
for this is that only sustained improvement would be likely to have an impact on
other critical outcomes, such as mental state, psychosocial functioning,
hospitalisation and relapse.

In the 2002 guideline, seven RCTs of cognitive remediation were included. Two trials
(Bellack2001 and Tompkins1995) were removed from the 2009 guideline analysis as
the GDG felt that they did not meet the definition of cognitive remediation. The
search for the 2009 guideline identified 15 papers providing follow-up data to
existing trials and 15 new trials. A recent meta-analysis (McGurk et al., 2007)
identified three additional trials and a number of other studies that did not meet
inclusion criteria. The cognitive remediation studies included in the trials employed
a variety of different methods and in some cases applied cognitive remediation in
combination with a variety of other psychological or psychosocial interventions 26. In
total, 25 trials (N = 1,390) met the inclusion criteria. All of the trials were published
in peer-reviewed journals between 1994 and 2008 (further information about both
9.5.4 Cognitive remediation versus control

For the 2009 guideline review, six of the included studies (Benedict1994;
BURDA1994; EACK2007; KURTZ2007; SATORY2005; VOLLEMA1995) did not
provide useable data for any of the critical outcomes listed in
Table 71. Consequently, 20 RCTs of cognitive remediation versus any type of control
were included in the meta-analysis (see Table 72 for a summary of the study
characteristics). Where there was sufficient data, sub- analyses were used to examine
cognitive remediation versus standard care and versus other active treatment. Forest
plots and/or data tables for each outcome can be found in Appendix 23d.

In the six RCTs (out of 17 included in the meta-analysis) that reported cognitive
outcomes at follow-up, there was limited evidence that cognitive remediation
produced sustained benefits in terms of cognition. However, these effects were
driven primarily by two studies (HOGARTY2004; PENADES2006); therefore,
sensitivity analyses were used to explore how robust the findings were. Removal of
these studies led to the loss of effects for all but one cognitive domain (reasoning and
problem solving). There was limited evidence suggesting that cognitive remediation
when compared with standard care may improve social functioning. However, this
effect was driven by a range of studies conducted by Velligan and colleagues
(VELLIGAN2000, 2002, 2008A, 2008B), in which the intervention was more
26Trials assessing the efficacy of cognitive remediation as an adjunct to non-psychological or
psychosocial interventions were outside the scope of the review. However, a review of cognitive
remediation with vocational rehabilitation interventions can be found in Chapter 13.

comprehensive than typical cognitive remediation programmes in the UK, and
included the use of individually tailored environmental supports to ameliorate areas
in addition to basic cognitive functions. The UK-based studies, although well-
conducted, did not report evidence of improvement in social or vocational
functioning or symptoms at either end of treatment or follow-up.
Table 71: Clinical review protocol for the review of cognitive remediation
Electronic databases Databases: CINAHL, CENTRAL, EMBASE, MEDLINE,

PsycINFO
Date searched Data base inception to 30July2008
Patient population Adults (18+) with schizophrenia (including schizophrenia-
related disorders)
Excluded populations Very late onset schizophrenia (onset after age 60) Other
Interventions Cognitive remediation
Quality of life
Cognitive outcomes (at follow-up only)a
Adverse events
aCognitive measures were categorised into the following cognitive domains based upon Nuechterlein
and colleagues, 2004: attention/vigilance, speed of processing, working memory, verbal learning and
memory, visual learning and memory, reasoning and problem solving, verbal comprehension, and
social cognition. The effect sizes for each individual measure were pooled to produce one effect size
per domain for each study.

Table 72: Summary of study characteristics for cognitive remediation
Cognitive remediation Cognitive remediation Cognitive remediation

versus any control versus standard care versus other active
treatments
k (total N) 17 (1084) 10 (522) 9 (605)
Study ID BELLUCCI2002 BELLUCCI2002 Hadaslidor2001
Hadaslidor2001 Medalia2000 HOGARTY2004
HOGARTY2004 SILVERSTEIN2005a Medalia1998
Medalia1998 TWAMLEY2008 PENADES2006
Medalia2000 VELLIGAN2000 SPAULDING1999
PENADES2006 VELLIGAN2002 VANDERGAAG2002
SILVERSTEIN2005a VELLIGAN2008A VELLIGAN2008A
SPAULDING1999 VELLIGAN2008B VELLIGAN2008B
TWAMLEY2008 WYKES2007A Wykes1999
VANDERGAAG2002 WYKES2007B
VELLIGAN2000
VELLIGAN2002
VELLIGAN2008A
VELLIGAN2008B
Wykes1999
WYKES2007A
WYKES2007B
Continued

Cognitive remediation Cognitive remediation Cognitive remediation

versus any control versus standard care versus other active
treatments
Diagnosis 83–100%schizophrenia 95–100% schizophrenia 83–100% schizophrenia
Orother related diagnoses Or other related diagnoses Or other related diagnoses
(DSM or ICD-10) (DSM or ICD-10) (DSM or ICD-10)
Baseline severity BPRS total: BPRS total: BPRS total:
Mean (SD) ~30 (4) Mean (SD) ~37 (9) Mean (SD)~30 (4)
Medalia1998 WYKES2007B Medalia1998
Mean (SD) ~37 (9)
WYKES2007B
PANSS total: PANSS total:
Mean (SD)~60 (15) Mean (SD) ~ 60 (15)
WYKES2007A WYKES2007A
Length of treatment Range: 5–104 weeks Range: 5–104 weeks Range: 6–104 weeks
Length of follow-up Up to 3 months: Up to 3 months: TWAMLEY2008
TWAMLEY2008 WYKES2007B
WYKES2007B Up to 6 months: Up to 6 months:
Up to 6 months: WYKES2007A PENADES2006
PENADES2006 Wykes1999
Wykes1999
WYKES2007A Up to 12 months:
Up to 12 months: HOGARTY2004
HOGARTY2004

Setting Inpatientb: Inpatientb: Inpatientb:
Medalia1998 Medalia2000 Medalia1998
Medalia2000 SILVERSTEIN2005 SPAULDING1999
SILVERSTEIN2005 WYKES2007B VANDERGAAG2002
SPAULDING1999
VANDERGAAG2002
WYKES2007B
Outpatient:
BELLUCCI2002 Outpatient: Outpatient:
HOGARTY2004 BELLUCCI2002 HOGARTY2004
VELLIGAN2000c VELLIGAN2008A
VELLIGAN2000c
VELLIGAN2002 VELLIGAN2008B Wykes1999
VELLIGAN2002
VELLIGAN2008A VELLIGAN2008A
VELLIGAN2008B
VELLIGAN2008B Wykes1999
WYKES2007A WYKES2007A
Day rehabilitation centre:
Hadaslidor2001 Day rehabilitation centre:
Hadaslidor2001
Note. aThe study included an attentional module for both cognitive remediation and waiting list control participants. The
attentional module started after the completion of the cognitive remediation intervention and after testing at time point two.
Only data from time point two were used in the analysis as this represented cognitive remediation versus standard care alone.
bIncluded inpatient rehabilitation units.
cParticipants in the Velligan papers were recruited following discharge from an inpatient setting.

Overall, there was no consistent evidence that cognitive remediation alone is
effective in improving the critical outcomes, including relapse rates,
rehospitalisation, mental state and quality of life. Furthermore, where effects of
treatment were found, the evidence is difficult to interpret as many studies report
non-significant findings without providing appropriate data for the meta-analysis.
Thus, the magnitude of the effect is likely to be overestimated for all outcomes.
9.5.6 Linking evidence to recommendation

The 2002 guideline found no consistent evidence for the effectiveness of cognitive
remediation versus standard care or any other active treatment in improving
targeted cognitive outcomes or other critical outcomes, such as symptom reduction.
It is noteworthy that although the McGurk and colleagues’ (2007) review suggested
positive effects for symptoms and functioning, this may be, in part, attributed to the
fact that their review included a number of studies that failed to meet the inclusion
criteria set out by the GDG (for example, minimum number of participants or
cognitive remediation as an adjunct to vocational rehabilitation).
Although limited evidence of efficacy has been found in a few recent well-
conducted studies, there is a distinct lack of follow-up data and various
methodological problems in the consistency with which outcomes are reported.
Where studies comprehensively reported outcomes at both ends of treatment and
follow-up, there was little consistent advantage of cognitive remediation over
standard care and attentional controls. Consequently, although there are some
positive findings, the variability in effectiveness suggests that the clinical evidence as
a whole is not robust enough to change the 2002 guideline.
The GDG did note, however, that a number of US-based studies have shown
sustained improvements in vocational and psychosocial outcomes when cognitive
remediation is added to vocational training and/or supported employment services.
Despite the emerging evidence within this context, the effectiveness of psychological
and psychosocial interventions as adjuncts to supported employment services was
outside the scope of the 2009 guideline and, therefore, has not been reviewed
systematically. Given this finding and the variability in both the methodological
rigour and effectiveness of cognitive remediation studies, it was the opinion of the
GDG that further UK-based research is required. In particular, RCTs of cognitive
remediation should include adequate follow-up periods to comprehensively assess
its efficacy as a discrete and/or adjunctive intervention.

9.5.7.1 An adequately powered RCT with longer-term follow-up should be
conducted to investigate the clinical and cost effectiveness of cognitive
remediation compared with an appropriate control in people with

9.6 COUNSELLING AND SUPPORTIVE THERAPY
9.6.1 Introduction
In the 1950s Carl Rogers, a pioneering US psychologist influenced by Alfred Adler
and Otto Rank, devised ‘client-centred’ and later ‘person-centred’ counselling. This
was a reaction against the behaviourist and psychodynamic schools that had
emerged from late 19th century Freudian psychoanalysis. Unlike the early
behaviourists, Rogers accepted the importance of a client’s internal emotional world,
but this centred on the lived experience of the person rather than empirically
untestable psychoanalytic theories of unconscious drives and defences of
unconscious processes (Thorne, 1992). Rogerian counselling has since been the
starting point for newer therapies, such as humanistic counselling, psychodynamic
counselling, psychodrama and Gestalt psychotherapy. In the UK, counselling is most
likely to be offered to people with common mental illnesses within a primary care
setting.
Supportive therapy has been cited as the individual psychotherapy of choice for
most patients with schizophrenia (Lamberti & Herz, 1995). It is notable that most
trials involving this intervention have used it as a comparison treatment for other
more targeted psychological approaches, rather than investigating it as a primary
intervention. This may be because supportive therapy is not a well-defined unique
intervention, has no overall unifying theory and is commonly used as an umbrella
term describing a range of interventions from befriending to a type of formal
psychotherapy (Buckley et al., 2007). More formal supportive therapy approaches
tend to be flexible in terms of frequency and regularity of sessions, and borrow some
components from Rogerian counselling (namely an emphasis on empathic listening
and ‘non-possessive warmth’). These may be called ‘supportive psychotherapy’ and
also tend to rely on an active therapist who may offer advice, support and
reassurance with the aim of helping the patient adapt to present circumstances
(Crown, 1988). This differs from the dynamic psychotherapist, who waits for
material to emerge and retains a degree of opacity to assist in the development of a
transference relationship.
Undoubtedly there are overlaps between counselling, supportive therapy and the
other psychotherapies; known as ‘non-specific factors’, these are necessary for the
development of a positive treatment alliance and are a prerequisite for any
psychological intervention to stand a chance of success (Roth et al., 1996). Many of
these factors are also part of high-quality ‘standard care’, as well as forming the key
elements of counselling and supportive therapy. Fenton and McGlashan (1997)
reported that a patient’s feeling of being listened to and understood is a strong
predictor of, for example, medication compliance. Also, according to McCabe and
Priebe (McCabe & Priebe, 2004), the therapeutic relationship is a reliable predictor of
patient outcome in mainstream psychiatric care.

Definition
Counselling and supportive therapy were defined as discrete psychological
interventions that:
• are facilitative, non-directive and/or relationship focused, with the
content largely determined by the service user, and
• do not fulfil the criteria for any other psychological intervention.

eligibility criteria used for this section of the guideline, can be found in Table 73. The
primary clinical questions can be found in Box 1. A new systematic search for
relevant RCTs published since the 2002 guideline was conducted for the 2009
guideline (further information about the search strategy can be found in Appendix
20).
Table 73: Clinical review protocol for the review of counselling and supportive
therapy

PsycINFO
Patient population Adults (18+) with schizophrenia (including

schizophrenia-related disorders)
Interventions Counselling and supportive therapy
Quality of life
Adverse events

In the 2002 guideline, 14 RCTs (N = 1,143) of counselling and supportive therapy
were included. Two studies included in the 2002 guideline (Levine1998;
Turkington2000) were excluded from the 2009 guideline review because of
inadequate numbers of participants. The search for the 2009 guideline identified four
papers providing follow-up data to existing trials and six new trials. In total, 18

RCTs (N = 1,610) met the inclusion criteria for the 2009 guideline. All were published
in peer-reviewed journals between 1973 and 2007 (further information about both
9.6.4 Counselling and supportive therapy versus control

For the 2009 guideline review, 17 RCTs of counselling and supportive therapy versus
any type of control were included in the meta-analysis. One included trial
(Donlon1973) did not provide any useable data for the analysis. Sub-analyses were
then used to examine counselling and supportive therapy versus standard care,
versus other active treatment and versus CBT 27 (see Table 74 for a summary of the
study characteristics). Forest plots and/or data tables for each outcome can be found
in Appendix 23d.

In 17 RCTs comprising 1,586 participants there was evidence to suggest that
counselling and supportive psychotherapy do not improve outcomes in
schizophrenia when compared with standard care and other active treatments, most
notably CBT. A subgroup analysis of counselling and supportive therapy versus
CBT favoured CBT for a number of outcomes including relapse. However, it must be
noted that in these studies, counselling and supportive therapy was used as
comparators to control primarily for therapist time and attention, and thus were not
the focus of the research.
Existing subgroup comparisons exploring the format of the intervention (group versus individual
27
sessions) was also updated. However, there was insufficient data to draw any conclusions based on
this subgroup. Please refer to Appendix 23d for the forest plots and/or data tables for all subgroup
comparisons conducted

Table 74: Summary of study characteristics for counselling and supportive therapy
Counselling and Counselling and Counselling and Counselling and

supportive therapy supportive therapy supportive therapy supportive therapy
versus any control versus standard care versus other active versus CBT
treatment
K (total N) 17 (1586) 2 (262)e 17 (1452) 9 (678)

Study ID Eckman1992 Tarrier1998 Eckman1992 Haddock1999
Falloon1981 Lewis2002a Falloon1981 Hogarty1997
Haddock1999 Haddock1999 Kemp1996
Herz2000 Herz2000 JACKSON2007
Hogarty1997 Hogarty1997 Lewis2002a
JACKSON2007 JACKSON2007 PINTO1999
Kemp1996 Kemp1996 Sensky2000
Lewis2002a Lewis2002a Tarrier1998
Marder1996 Marder1996 VALMAGGIA2005
PATTERSON2006 PATTERSON2006
PINTO1999 PINTO1999
ROHRICHT2006 ROHRICHT2006
Sensky2000 Sensky2000
SHIN2002 SHIN2002
Stanton1984 Stanton1984
Tarrier1998 Tarrier1998
VALMAGGIA2005 VALMAGGIA2005
Diagnosis 58–100% schizophrenia or 88–98% schizophrenia or 58–100% schizophrenia or 58–100% schizophrenia or
other related diagnoses other related diagnoses other related diagnoses other related diagnoses
(DSM or ICD-10) (DSM or ICD-10) (DSM or ICD-10) (DSM or ICD-10)

Baseline severity BPRS total: PANSS total: BPRS total: BPRS total:
Mean (SD) range: Mean (SD) ~87 (17) Mean (SD) range: Mean (SD) range:
~32 (8) to ~92 (8) Lewis2000 ~32 (8) to ~92 (8) ~32 (8) to ~92 (8)
PANSS total: PANSS total: PANSS total:

Mean (SD) range: Mean (SD) range: Mean (SD) range:
~61 (27) to ~87 (17) ~61 (27) to ~87 (17) ~61 (27) to ~87 (17)
CPRS total: CPRS total: CPRS total:

Mean (SD) ~36 (14) Mean (SD) ~36 (14) Mean (SD) ~36 (14)
Sensky2000 Sensky2000 Sensky2000
Length of Range: 5 to 156 weeks Range: 5 to 10 weeks Range: 5 to 156 weeks Range: 5 to 156 weeks
treatment
Length of follow- up Range: 4 to 24 months Range: up to 24 months Range: 4 to 156 months Range: 4 to 24 months
(only including
papers reporting
follow-up
measures)
Continued

Counselling and Counselling and Counselling and Counselling and

supportive therapy supportive therapy supportive therapy supportive
versus any control versus standard care versus other active therapy versus CBT
treatment
Haddock1999 Lewis2002c Haddock1999 Haddock1999
Hogarty1997b Hogarty1997b Hogarty1997b
Kemp1996 Kemp1996 Lewis2002c
Lewis2002c Lewis2002c VALMAGGIA2005
Stanton1984 Stanton1984
VALMAGGIA2005 VALMAGGIA2005
Outpatient: Outpatient: Outpatient: Outpatient:

Falloon1981 Tarrier1998 Falloon1981 Sensky2000
Herz2000 Herz2000 Tarrier1998
Marder1996 Marder1996
ROHRICHT2006 ROHRICHT2006
SHIN2002 SHIN2002
Sensky2000 Sensky2000
Tarrier1998 Tarrier1998
Eckmann1992 Eckmann1992 PINTO1999
PINTO1999 PINTO1999
Otherd: Otherd: Otherd:
JACKSON2007 JACKSON2007 JACKSON2007
PATTERSON2006 PATTERSON2006
Note. aFollow-up papers to Lewis2002 report the data separately for the three study sites, hence in the analysis Lewis2002appears
as LEWIS2002L (Liverpool), LEWIS2002M (Manchester) and LEWIS2002N (Nottingham).
bParticipants were recruited in the inpatient setting with the interventions starting shortly before discharge.
cParticipants were recruited from inpatient wards and day hospitals.
dOther settings included Board and Care facilities and EIS settings.
eBoth studies included multiple treatment arms; only the numbers in the counselling and supportive therapy and standard care
arms have been included in this count.

In the 2002 guideline, the GDG found no clear evidence to support the use of
counselling and supportive therapy as a discrete intervention. The limited evidence
found for the 2009 guideline does not justify changing this recommendation. The
GDG does, however, acknowledge the preference that some service users and carers
may have for these interventions, particularly when other more efficacious
psychological treatments are not available in the local area. Furthermore, the GDG
recognise the importance of supportive elements in the provision of good quality
standard care.
9.6.7 Recommendation
9.6.7.1 Do not routinely offer counselling and supportive psychotherapy (as specific
interventions) to people with psychosis or schizophrenia. However, take
service user preferences into account, especially if other more efficacious
psychological treatments, such as CBT, family intervention and arts
therapies, are not available locally. [2009]
9.7 FAMILY INTERVENTION

9.7.1 Introduction
Family intervention in the treatment of schizophrenia has evolved from studies of
the family environment and its possible role in affecting the course of schizophrenia
(Vaughn & Leff, 1976) after an initial episode. It should be noted that in this context,
‘family’ includes people who have a significant emotional connection to the service
user, such as parents, siblings and partners. Brown and colleagues (Brown et al.,
1962; Brown & Rutter, 1966) developed a measure for the level of ‘expressed
emotion’ within families and were able to show that the emotional environment
within a family was an effective predictor of relapse in schizophrenia (Bebbington &
Kuipers, 1994; Butzlaff & Hooley, 1998) The importance of this work lay in the
realisation that it was possible to design psychological methods (in this case, family
intervention) that could change the management of the illness by service users and
their families, and influence the course of schizophrenia.
Family intervention in schizophrenia derives from behavioural and systemic ideas,

adapted to the needs of families of those with psychosis. More recently, cognitive
appraisals of the difficulties have been emphasised. Models that have been
developed aim to help families cope with their relatives’ problems more effectively,
provide support and education for the family, reduce levels of distress, improve the
ways in which the family communicates and negotiates problems, and try to prevent
relapse by the service user. Family intervention is normally complex and lengthy
(usually more than ten sessions) but delivered in a structured format with the
individual family, and tends to include the service user as much as possible.

Definition
Family intervention was defined as discrete psychological interventions where:
• family sessions have a specific supportive, educational or treatment
function and contain at least one of the following components:
- problem solving/crisis management work, or
- intervention with the identified service user.

relevant RCTs published since the 2002 guideline was conducted for the 2009
20 and information about the search for health economic evidence can be found in
Section 9.7.8).
Table 75: Clinical review protocol for the review of family intervention

PsycINFO
Date searched 1January 2002 to 30 July 2008
Study design RCT (≥10 participants per arm and ≥ 6weeks’
duration)
Patient population Adults (18+) with schizophrenia (including schizophrenia-
related disorders)
Interventions Family intervention
Global state (relapse, rehospitalisation,)
Family outcomes (including burden)
Quality of life
Adverse events

In the 2002 guideline, 18 RCTs (N = 1,458) of family intervention were included. One
study (Posner1992) included in the 2009 guideline was re-classified as
‘psychoeducation’ for the 2009 guideline and two previous trials were classified as
having family intervention as part of a multi-modal treatment (Herz2000 and

Lukoff1986). The search for the 2009 guideline identified five papers providing
follow-up data to existing trials and 19 new trials. In total, 38 trials (N = 3,134) met
the inclusion criteria for the 2009 guideline review. All were published in peer-
reviewed journals between 1978 and 2008 (further information about both included
and excluded studies can be found in Appendix 22c).
9.7.4 Family intervention versus control

For the 2009 guideline, one of the included studies (CHENG2005) did not provide
useable data for any of the critical outcomes listed in Table 75, thus 32 RCTs of
family intervention versus any type of control were included in the meta-analysis.
Of these, 26 trials compared family intervention with standard care and eight
compared family intervention with other active treatments. Additionally, five trials
directly compared a multiple family intervention with a single family intervention
(see Table 76 for a summary of the study characteristics). Forest plots and/or data
tables for each outcome can be found in Appendix 23d.
Subgroup analyses were also used to examine whether the format of the family
intervention had an impact on outcome (ten trials were included in the analysis of
multiple family interventions versus any control and 11 trials were included in the
analysis of single family interventions versus any control). Additional subgroup
analyses were used to explore certain characteristics of the trials, such as the
inclusion of the person with schizophrenia, patient characteristics and the length of
the intervention28 (see Table 77 for a summary of the studies included in each
subgroup comparison).
9.7.5 Training
Although there was a paucity of information on training and/or competence of the
therapists in the RCTs of family intervention, 28 trials reported the profession of the
therapist. In these trials, the professional background varied, with the most
commonly reported professions being clinical psychologist (14/28) or psychiatric
nurse (12/28). In addition, the following professionals also conducted the
intervention in a number of papers: psychiatrist (10/28), social workers (3/28),
Masters’ level psychology graduates (2/28) and local mental health workers (1/28).
In many trials a number of therapists, often across different disciplines, conducted
the interventions, with some trials emphasising collaboration between the therapists
and the participant’s key worker.
28Existing subgroup comparisons exploring the country of the trial, the number of treatment sessions, and the
family characteristics (high emotional expression versus everything) were also updated. However, there was
insufficient data to draw any conclusions based on these subgroups. Please refer to Appendix 23d for the forest
plots and/or data tables for all subgroup comparisons conducted.

Table 76: Summary of study characteristics for family intervention
Family intervention versus any Family intervention Family intervention versus Multiple family versus
control versus standard care other active treatments single family
intervention (direct
format comparison)
K (total N) 32 (2429) 26 (1989) 8 (417) 5 (641)
Study ID Barrowclough1999 Barrowclough1999 CARRA2007 Leff1989
Bloch1995 Bloch1995 Falloon1981 McFarlane1995a
BRADLEY2006 BRADLEY2006 GARETY2008a McFarlane1995b
BRESSI2008 BRESSI2008 Herz2000b MONTERO2001
Buchkremer1995 Buchkremer1995 Hogarty1997 Schooler1997
CARRA2007 CARRA2007 LINSZEN1996b
CHIEN2004A CHIEN2004A Lukoff1986b SZMUKLER2003
CHIEN2004B CHIEN2004B
CHIEN2007 CHIEN2007
Dyck2000 Dyck2000
Falloon1981 GARETY2008a
GARETY2008a Glynn1992
Glynn1992 Goldstein1978
Goldstein1978 JENNER2004b
Herz2000b KOPELOWICZ2003
Hogarty1997 LEAVEY2004

Family intervention versus any Family intervention Family intervention versus Multiple family versus
control versus standard care other active treatments single family
intervention (direct
format comparison)
JENNER2004b Leff1982
KOPELOWICZ2003 LI2005
LEAVEY2004 MAGLIANO2006
Leff1982 RAN2003
LI2005 SO2006
LINSZEN1996b Tarrier1988
Lukoff1986b VALENCIA2007b
MAGLIANO2006 Vaughan1992
RAN2003 Xiong1994
SO2006 Zhang1994
SZMUKLER2003
Tarrier1988
VALENCIA2007b
Vaughan1992
Xiong1994
Zhang1994
Diagnosis 93–100% 93–100% 98–100% schizophrenia 100% schizophrenia
schizophrenia or schizophrenia or other related or other related
other related or other diagnoses (DSM or ICD- diagnoses (DSM or
diagnoses (DSM or related 10) ICD-10)
ICD-10) diagnoses
(DSM or ICD-
10)

Mean (SD) range: ~27 (3) Mean (SD) range: Mean (SD): 29 (7)
to ~48 (10) ~27 (3) to~48 (10) Schooler1997

Mean (SD) range:~53 (1) Mean (SD) range: Mean (SD) range:
To 112 (26) ~60 (14) to 112 (26) ~53 (17) to ~67 (14)
Length of Range: 6–156 weeks Range: 12–104 weeks Range: 6–156weeks Range: 52–104 weeks
treatment
Length of follow- Range: 3–60 months Range: 3–60 months Range: 12–60months Range: 24–60 months
up (only
including papers
reporting follow-
up measures)

Bloch1995c Bloch1995c Hogarty1997d Leff1989
BRESSI2008 BRESSI2008 LINSZEN1996b McFarlane1995a
Glynn1992 Glynn1992 Lukoff1986b
Hogarty1997d Vaughan1992
LINSZEN1996b
Lukoff1986b
Vaughan1992

Family intervention Family intervention Family intervention Multiple family versus

versus any control versus standard care versus other active single family
treatments intervention (direct
format comparison)
Outpatient: Outpatient: Outpatient: Outpatient:

Barrowclough1999 Barrowclough1999 CARRA2007 McFarlane1995b
BRADLEY2006 BRADLEY2006 Falloon1981 MONTERO2001
Buchkremer1995 Buchkremer1995 Herz2000b Schooler1997
CARRA2007 CARRA2007 SZMUKLER2003
CHIEN2004A CHIEN2004A
CHIEN2004B CHIEN2004B
CHIEN2007 CHIEN2007
Dyck2000 Dyck2000
Falloon1981 Goldstein1978e
Goldstein1978e Herz2000b JENNER2004b
JENNER2004b KOPELOWICZ2003
KOPELOWICZ2003 Leff1982
MAGLIANO2006

Leff1982 RAN2003
MAGLIANO2006 SO2006
RAN2003 Tarrier1998
SO2006 VALENCIA2007b
SZMUKLER2003 Xiong1994
Tarrier1998 Zhang1994
VALENCIA2007b
Xiong1994
Zhang1994

GARETY2008a GARETY2008a GARETY2008a
LEAVEY2004 LEAVEY2004
LI2005 LI2005
Note. Studies were categorised as short (12weeks or fewer), medium (12–51weeks) and long (52 weeks or more).
aOnly the carer pathway was included in the present analysis.
bMulti-modal interventions.
cCarers of patients admitted to the ward were recruited to take part in the study.
dParticipants were recruited in the inpatient setting with the intervention starting shortly before discharge.
eParticipants were recruited following discharge to an after care outpatient programme

Table 77: Summary of study characteristics for family intervention subgroup comparisons
Single family Multiple family Family intervention Family intervention

intervention versus any intervention versus any including service user excluding service user
control control versus any control Versus any control
K (total N) 11 (864) 10 (651) 18 (1319) 9 (622)

Study ID Barrowclough1999 BRADLEY2006 Barrowclough1999 Bloch1995
Bloch1995 Buchkremer1995 BRADLEY2006 Buchkremer1995
BRESSI2008 CARRA2007 BRESSI2008 CARRA2007
Falloon1981 CHIEN2004A CHIEN2004B CHIEN2004A
GARETY2008 CHIEN2004B CHIEN2007 Dyck2000
Glynn1992 CHIEN2007 Falloon1981 LEAVEY2004
Hogarty1997 Dyck2000 GARETY2008 SO2006
LEAVEY2004 KOPELOWICZ2003 Glynn1992 SZMUKLER2003
MAGLIANO2006 SO2006 Goldstein1978 Vaughan1992
RAN2003 Xiong1994 Hogarty1997
Vaughan1992 KOPELOWICZ2003
Leff1982
LI2005
MAGLIANO2006
RAN2003
Tarrier1988
Xiong1994
Zhang1994

Short-term family Medium-term family Long-term family

intervention versus intervention versus intervention versus
any control any control any control
K (total N) 4 (248) 12 (1056) 10 (660)

Study ID Bloch1995 Barrowclough1999 BRADLEY2006
Goldstein1978 CHIEN2004A BRESSI2008
SO2006 CHIEN2004B Buchkremer1995
Vaughan1992 CHIEN2007 CARRA2007
GARETY2008 Dyck2000
KOPELOWICZ2003 Falloon1981
LEAVEY2004 Glynn1992
Leff1982 Hogarty1997
MAGLIANO2006 Xiong1994
RAN2003 Zhang1994
SZMUKLER2003
Tarrier1988
Family intervention Family intervention Family intervention
versus any control– versus any control– versus any control–
first episodea acute episode promoting recovery
K (total N) 4 (333) 12 (673) 9 (702)
Study ID Goldstein1978 Bloch1995 Barrowclough1999
LEAVEY2004 BRADLEY2006 Buchkremer1995
SO2006 BRESSI2008 CARRA2007
Zhang1994 Falloon1981 CHIEN2004A
GARETY2008 CHIEN2004B
Glynn1992 CHIEN2007
Hogarty1997 Dyck2000
KOPELOWICZ2003 LI2005
Leff1982 MAGLIANO2006
Tarrier1988
Vaughan1992
Xiong1994
Note. aA number of trials included participants across different phases of illness (for
example, first episode, acute and promoting recovery) and hence could not be included
in the subgroup analysis.

9.7.6 Ethnicity
Although the data on ethnicity was limited, a subgroup analysis looking at the
efficacy of family intervention in an ethnically diverse population was conducted
(see Chapter 6 for definition of ethnically diverse sample). For critical outcomes
including relapse, rehospitalisation and symptoms, family intervention was shown
to have clinically significant benefits within studies including an ethnically diverse
sample. One UK study (LEAVEY2004) assessed the impact of a brief family
intervention for families of patients with first episode psychosis. Participants were
drawn from a multicultural and ethnically diverse population, with the researchers
attempting to match the ethnicity of the family worker with the ethnicity of the carer.
LEAVEY2004 failed to demonstrate any significant impact on ether patient outcomes
or carer level of satisfaction. However, the authors note that the high proportion
failing to take up the intervention may have had a detrimental impact upon the
results.
A number of papers have assessed the effectiveness of adapting a Western family

intervention approach to better suit non-Western populations. For example, both
RAN2003 and LI2005 adapted the content of the intervention to better match the
cultural needs and family structures of people living in different communities in
mainland China. Further to this, researchers have started to assess the impact of
cultural modifications aimed at tailoring an intervention to better suit the cultural
and ethnic needs of minority populations. For instance, BRADLEY2006 assessed the
effectiveness of a modified intervention approach that included the use of language
matching and ethno-specific explanatory models in a sample of Vietnamese
speaking migrants living in Australia. Although both types of cultural modifications
were shown to be effective across critical outcomes, none of the RCTs was conducted
with black and minority ethnic participants from the UK; therefore the
generalisability of such findings is limited. Furthermore, at present little research
exists that directly compares the efficacy and acceptability of culturally and non-
culturally modified approaches.

In 32 RCTs including 2,429 participants, there was robust and consistent evidence for
the efficacy of family intervention. When compared with standard care or any other
control, there was a reduction in the risk of relapse with numbers needed to treat
(NNTs) of 4 (95% CIs 3.23 to 5.88) at the end of treatment and 6 (95% CIs. 3.85 to
9.09) up to 12 months following treatment. In addition, family intervention also
reduced hospital admission during treatment and the severity of symptoms both
during and up to 24 months following the intervention. Family intervention may
also be effective in improving additional critical outcomes, such as social functioning
and the patient’s knowledge of the disorder. However, it should be noted that
evidence for the latter is more limited and comes from individual studies reporting
multiple outcomes across a range of scale-based measures.

The subgroup analyses conducted for the 2009 guideline to explore the variation in
terms of intervention delivery consistently indicated that where practicable the
service user should be included in the intervention. Although direct format
comparisons did not indicate any robust evidence for single over multiple family
intervention in terms of total symptoms, single family intervention was seen as more
acceptable to service users and carers as demonstrated by the numbers leaving the
study early. Additionally, subgroup comparisons that indirectly compared single
with multiple family intervention demonstrated some limited evidence to suggest
that only the former may be efficacious in reducing hospital admission.

No studies evaluating the cost effectiveness of family intervention for people with
schizophrenia met the set criteria for inclusion in the guideline systematic review of
economic literature. However, the 2002 guideline, using more relaxed inclusion
criteria, had identified a number of economic studies on family intervention for
people with schizophrenia. Details on the methods used for the systematic search of
the economic literature in the 2009 guideline are described in Appendix 11. The
following text marked by asterisks is derived from the 2002 guideline.
**2002** The economic review identified five eligible studies, and a further two
studies were not available. All five included studies were based on RCTs. Three
papers adapted simple costing methods (Goldstein, 1996; Leff, 2001; Tarrier et al.,
1991), while two studies were economic evaluations (Liberman et al., 1987;
McFarlane et al., 1995). Of these, two economic analyses were conducted in the UK
(Leff, 2001; Tarrier et al., 1991) and two others were based on clinical data from the
UK, but the economic analyses were conducted within a US context (Goldstein, 1996;
Liberman et al., 1987). Most of these studies are methodologically weak, with the
potential for a high risk of bias in their results. Another common problem was the
low statistical power of the studies to show cost differences between the
comparators. All studies focused narrowly on direct medical costs. As such,
economic evaluation of family interventions from a broader perspective is
impossible.
One study (Tarrier et al., 1991) compared family intervention with standard care and
concluded that family intervention is significantly less costly than standard care.
Two analyses compared family intervention with individual supportive therapy
(Goldstein, 1996; Liberman et al., 1987). Both studies used clinical data from the same
RCT, but their evaluation methodology differed. They concluded that the treatment
costs of family intervention are higher than those of individual supportive therapy,
but cost savings relating to other healthcare costs offset the extra treatment costs.
One study (Leff, 2001) showed economic benefits of family intervention combined
with two psychoeducational sessions over psychoeducation alone. However, the
difference was not significant. One study (McFarlane et al., 1995) demonstrated that

multi- family group intervention is more cost effective than single-family
intervention.
The quality of the available economic evidence is generally poor. The evidence, such
as it is, suggests that providing family interventions may represent good ‘value for
money’. There is limited evidence that multi-family interventions require fewer
resources and are less costly than single-family interventions. **2002**
The evidence table for the above studies as it appeared in the 2002 guideline is
included in Appendix 25.
Economic modelling
Objective
**2009**The guideline systematic review and meta-analysis of clinical evidence
demonstrated that provision of family intervention is associated with a reduction in
relapse and hospitalisation rates of people with schizophrenia. A cost analysis was
undertaken to assess whether the costs of providing family intervention for people
with schizophrenia are offset by cost savings to the NHS following this decrease in
relapse and hospitalisation rates.
Intervention assessed
Family intervention can be delivered to single families or in groups. The guideline
meta-analysis included all studies of family intervention versus control in its main
analysis, irrespective of the mode of delivery, because it was difficult to distinguish
between single and multiple programmes. The majority of studies described family
intervention programmes that were predominantly single or multiple, but might
have some multiple or single component, respectively; some of the interventions
combined single and multiple sessions equally.
Apart from the main meta-analysis, studies of family intervention versus control
were included in additional sub-analyses in which studies comparing
(predominantly) single family intervention versus control were analysed separately
from studies comparing (predominantly) multiple family intervention versus
control. These sub-analyses demonstrated that single family intervention
significantly reduced the rates of hospital admission of people with schizophrenia
up to 12 months into therapy, whereas multiple family intervention was not
associated with a statistically significant respective effect. On the other hand, single
and multiple family intervention had a significant effect of similar magnitude in
reducing the rates of relapse.
A small number of studies compared directly (exclusively) single with (exclusively)

multiple family intervention. Meta-analysis of these studies showed that single and
multiple family intervention had no significant difference in clinical outcomes.
However, participants showed a clear preference for single interventions, as
expressed in dropout rates.

It was decided that the economic analysis would utilise evidence from the main
meta-analysis of all studies on family intervention versus control (irrespective of the
model of delivery) but, in terms of intervention cost, would consider single family
intervention; this would produce a conservative cost estimate per person with
schizophrenia, given that in multiple family intervention the intervention cost is
spread over more than one family.
Methods
A simple economic model estimated the total net costs (or cost savings) to the NHS
associated with provision of single family therapy, in addition to standard care, to
people with schizophrenia and their families/carers. Two categories of costs were
assessed: costs associated with provision of family intervention, and cost savings
from the reduction in relapse and hospitalisation rates in people with schizophrenia
receiving family intervention, estimated based on the guideline meta-analysis of
respective clinical data. Standard care costs were not estimated because these were
common to both arms of the analysis.
Cost data
Intervention costs (costs of providing family intervention) The single family intervention
programmes described in the clinical studies included in the guideline systematic
review were characterised by a wide variety in terms of number of sessions and
duration of each session. The resource use estimate associated with provision of
single family intervention in the economic analysis was based on the expert opinion
of the GDG regarding optimal clinical practice in the UK, and was consistent with
average resource use reported in these studies. Single family intervention in the
economic analysis consisted of 20 hours and was delivered by two therapists.
As with CBT, the GDG acknowledge that family intervention programmes can be
delivered by a variety of mental health professionals with appropriate training and
supervision. The salary level of a mental health professional providing family
intervention was estimated to be similar to that of a mental health professional
providing CBT, and comparable with the salary level of a clinical psychologist.
Therefore, the unit cost of a clinical psychologist was used to estimate an average
intervention cost. The unit cost of a clinical psychologist is estimated at £67 per hour
of client contact in 2006/07 prices (Curtis, 2007). This estimate is based on the mid-
point of Agenda for Change salaries Band 7 of the April 2006 pay scale, according to
the National Profile for Clinical Psychologists, Counsellors and Psychotherapists
(NHS Employers, 2006). It includes salary, salary oncosts, overheads and capital
overheads, but does not take into account qualification costs because the latter are
not available for clinical psychologists.
Based on the above resource use estimates and the unit cost of a clinical
psychologist, the cost of providing a full course of family intervention was estimated
at £2,680 per person with schizophrenia in 2006/07 prices.

Costs of hospitalisation/cost-savings from reduction in hospitalisation rates As described in
Section 9.4.8, the average cost of hospitalisation per person with schizophrenia was
estimated at £28,645 in 2006/07 prices, based on national statistics on the mean
length of hospitalisation for people with schizophrenia (NHS, The Information
Centre, 2008a) and the NHS reference cost per bed-day of an inpatient mental health
acute care unit for adults, in 2006/07 prices (Department of Health, 2008).
Clinical data on hospitalisation rates following provision of family intervention

The guideline meta-analysis provided pooled data on both hospitalisation and
relapse rates associated with provision of family intervention in addition to standard
care versus standard care alone. The analyses showed that adding family
intervention to standard care significantly reduced the rates of both hospitalisation
and relapse in people with schizophrenia. The vast majority of these data came
from studies conducted outside the UK. The GDG expressed the view that
hospitalisation levels may differ significantly across countries, depending on
prevailing clinical practice, and therefore data on hospitalisation rates derived from
non-UK countries might not be applicable to the UK setting. On the other hand, the
definition of relapse was more consistent across studies (and countries). For this
reason, it was decided to use pooled data on relapse rather hospitalisation rates for
the economic analysis; these data would be used, subsequently, to estimate
hospitalisation rates relevant to people with schizophrenia in the UK to calculate
cost savings from reducing hospital admissions following provision of family
intervention.
The guideline meta-analysis of family intervention data on relapse rates included

two analyses: one analysis explored the effect on relapse rates during treatment with
family intervention, and another analysis estimated the effect on relapse rates at
follow-up, between 4 and 24 months after completion of family intervention. Ideally,
both analyses should be taken into account at the estimation of total savings
associated with family intervention. However, follow-up data were not
homogeneous: some studies reported relapse data during treatment separately from
respective data after treatment, but other studies included events that occurred
during treatment in the reported follow-up data. Taking into account both sets of
data might therefore double-count events occurring during treatment and would
consequently overestimate the value of cost savings associated with family
intervention. It was decided to use relapse data during treatment in the analysis,
because these data were homogeneous and referred to events that occurred within
the same study phase. It is acknowledged, however, that the cost savings estimated
using data exclusively reported during treatment are probably underestimates of the
true cost savings because the beneficial effect of family intervention on relapse
remains for a substantial period after completing treatment.
Table 78 shows the family intervention studies included in the meta-analysis of

relapse rate data for 1 to 12 months into treatment, the relapse rates for each
treatment arm reported in the individual studies and the results of the meta-analysis.

The results of the meta-analysis show that family intervention, when added to
standard care, reduces the rate of relapse in people with schizophrenia during the
intervention period (the RR of relapse of family intervention added to standard care
versus standard care alone is 0.52). This result was significant at the 0.05 level (95%
CIs of RR: 0.42 to 0.65). It must be noted that the meta-analysis of relapse follow-up
data showed that this beneficial effect remains significant up to at least 24 months
after the end of therapy (respective RR up to 24 months following provision of
family intervention 0.63, with 95% CIs 0.52 to 0.78).
Table 78: Studies considered in the economic analysis of family intervention

added to standard care versus standard care alone and results of the meta-analysis
(1 to 12 months into treatment)
Study ID Total events (n) in each treatment arm (N)

Family intervention Standard care alone
plus standard care (n/N) (n/N)
GOLDSTEIN1978 7/52 12/52

LEFF1982 1/12 6/12
TARRIER1988 13/32 20/32
GLYNN1992 3/21 11/20
XIONG1994 12/34 18/29
BARROWCLOUGH1999 9/38 18/39
RAN2003 22/57 32/53
BRADLEY2006 8/30 13/29
BRESSI2008 3/20 13/20
TOTAL 78/296 (26.35%) 143/286 (50.00%)
Meta-analysisresults RR: 0.52 95% CI: 0.42–0.65
The baseline rate of relapse in the economic analysis was taken from the overall rate
of relapse under standard care alone, as estimated in the guideline meta-analysis of
family intervention data on relapse; that is, a 50% baseline relapse rate was used. The
rate of relapse when family intervention was added to standard care was calculated
by multiplying the estimated RR of relapse of family intervention plus standard care
versus standard care alone by the baseline relapse rate.
Details on the studies considered in the economic analysis are available in Appendix
22c. The forest plots of the respective meta-analysis are provided in Appendix 23d.
Association between relapse and hospitalisation rates

In the UK, people with schizophrenia experiencing a relapse are mainly treated
either as inpatients or by CRHTTs. Glover and colleagues (2006) examined the
reduction in hospital admission rates in England following the implementation of

CRHTTs. They reported that the introduction of CRHTTs was followed by a 22.7%
reduction in hospital admission levels. Based on this data, the economic analysis
assumed that 77.3% of people with schizophrenia experiencing a relapse would be
admitted in hospital, and the remaining 22.7% would be seen by CRHTTs.
One- and two-way sensitivity analyses were undertaken to investigate the
robustness of the results under the uncertainty characterising some of the input
parameters and the use of different assumptions in the estimation of total net costs
(or net savings) associated with provision of family intervention for people with
schizophrenia. The following scenarios were explored:
• Use of the 95% CIs of the RR of relapse of family intervention added to
standard care versus standard care alone.
• Change in the total number of hours of a course of family intervention
(20 hours in the base-case analysis) to between a range of 15 and 25
hours.
• Change in the baseline rate of relapse (that is, the relapse rate for
standard care) from 50% (that is, the baseline relapse rate in the base-
case analysis) to a more conservative value of 30%.
• Change in the rate of hospitalisation following relapse (77.3% in base-
case analysis) to 61.6% (based on the upper 95% CI of the reduction in
hospital admission levels following the introduction of CRHTTs which,
according to Glover and colleagues (2006), was 38.4%).
• Simultaneous use of a 30% relapse rate for standard care and a 61.6%
hospitalisation rate following relapse.
• Use of a lower value for duration of hospitalisation. A value of 69 days
was tested, taken from an effectiveness trial of clozapine versus SGAs
conducted in the UK (CUtLASS Band 2, (Davies et al., 2008).
Results
Base-case analysis Providing family intervention cost £2,680 per person. The reduction
in the rates of relapse in people with schizophrenia during treatment with family
intervention in addition to standard care resulted in cost savings equaling
£5,314 per person. Thus, family intervention resulted in an overall net saving of
£2,634 per person with schizophrenia. Full results of the base-case analysis are
reported in Table 79.
Table 79: Results of cost analysis comparing family intervention in addition to

standard care with standard care alone per person with schizophrenia
Costs Family intervention plus Standard care alone Difference

standard care
Family intervention cost £2,680 0 £2,680
Hospitalisation cost £5,757 £11,071 −£5,314
Totalcost £8,437 £11,071 −£2,634

Sensitivity analysis The results of the base-case analysis were overall found to be
robust to the different scenarios explored in sensitivity analysis. Family intervention
remained cost saving when the 95% CIs of the RR of relapse during treatment were
used. In most scenarios, using the mean RR of relapse taken from the guideline
meta-analysis, the addition of family intervention to standard care resulted in overall
cost savings because of a substantial reduction in relapse and subsequent
hospitalisation costs. The only scenario in which family intervention was not cost
saving (instead incurring a net cost of £139 per person) was when a 30% baseline
relapse rate was assumed, combined with a 61.6% rate of hospitalisation following
relapse (in this scenario, the overall cost ranged between a net saving of £390 and a
net cost of £827 when the 95% CIs of RR of relapse were used). Full results of
sensitivity analysis are presented in Table 80.
Discussion
The economic analysis showed that family intervention for people with
schizophrenia is likely to be an overall cost-saving intervention because the
intervention costs are offset by savings resulting from a reduction in the rate of
relapses experienced during therapy. The net cost saving of providing family
intervention ranged between £1,195 and £3,741 per person with schizophrenia, using
a mean duration of hospitalisation of 110.6 days and the 95% CIs of RRs of relapse,
as estimated in the guideline meta-analysis. When a mean length of hospital stay of
69 days was used, the net cost of providing family intervention was found to lie
between −£1,326 (overall net saving) and £263 per person with schizophrenia.
Table 80: Results of sensitivity analysis of providing family intervention in

addition to standard care for people with schizophrenia
Scenario Total net cost (negative cost implies net

saving)
Use of 95%CIs of RR of relapse −£3,741 (lower CI) to −£1,195
(upper CI)
Family intervention hours between 15 and −£3,304 to −£1,964 respectively

25
Relapse rate under standard care30% −£509 (−£1,173 to £355 using the
95%CIs of RR of relapse)
Rate of hospitalisation following relapse −£1,555 (−£2,437 to −£408 using the

61.6% 95%CIs of RR of relapse)
Relapse rate under standard care30% and £139 (−£390 to £827 using the 95%
rate of hospitalisation following relapse CIs of RR of relapse)
61.6%
Mean length of hospitalisation 69 days −£635 (−£1,326 to £263 using the
95%CIs of RR of relapse)

The economic analysis estimated cost savings related exclusively to a decrease in
hospitalisation costs following reduction in relapse rates associated with family
intervention. Consideration of further potential cost savings, such as savings
resulting from an expected reduction in contacts with CRHTTs following reduction
in relapse rates, would further increase the cost savings associated with family
intervention. Moreover, meta-analysis of follow-up data demonstrated that the
beneficial effect of family intervention on relapse rates observed in people with
schizophrenia remains significant for a period at least 24 months following
treatment. This means that the cost savings associated with family intervention are
even higher. Finally, the expected improvement in HRQoL of people with
schizophrenia and their carers following a reduction in relapse rates further
strengthens the argument that family intervention is likely to be a cost-effective
option for people with schizophrenia in the UK.

There was sufficient evidence in the 2002 guideline for the GDG to recommend
family intervention in the treatment of schizophrenia. Recent studies have
corroborated these conclusions and have consistently shown that family intervention
may be particularly effective in preventing relapse.
Further analyses undertaken for the 2009 guideline continue to support the evidence
demonstrated in the 2002 guideline with regard to the duration of treatments and
the inclusion of the person with schizophrenia, where practicable. Although the
evidence is more limited for the advantages of single compared with multiple family
interventions, this must be considered in the context of current practice as well as
service user and carer preferences. Furthermore, the GDG noted that the majority of
UK-based studies were conducted as single family interventions, with the non-UK
studies contributing more to the multiple family intervention evidence base. Thus,
the evidence for single family intervention may additionally be more generalisable
to UK settings.
Existing economic evidence on family intervention is poor. A simple economic

analysis undertaken for this guideline demonstrated that, in the UK setting, family
intervention is associated with net cost savings when offered to people with
schizophrenia in addition to standard care, owing to a reduction in relapse rates and
subsequent hospitalisation. The findings of the economic analysis used data on
relapse that referred to the period during treatment with family intervention.
However, there is evidence that family intervention also reduces relapse rates for a
period after completion of the intervention. Therefore, net cost savings from family
intervention are probably higher than those estimated in the guideline economic
analysis.
With regard to the training and competencies required by the therapist to deliver
family intervention to people with schizophrenia and their carers, there was a
paucity of information reported throughout the trials. Consequently, the GDG were
unable to form any conclusions or make any recommendations relating to practice.

However, the GDG acknowledges that the training and competencies of the
therapist is an important area, and one that warrants further research.
The robust evidence presented in the current clinical and health economic evaluation
of family intervention further supports the conclusions and recommendations in the
2002 guideline. Although there was a lack of evidence for the use of culturally
adapted family interventions within the UK, the GDG acknowledges that this is an
important area warranting further investigation given the evidence previously
discussed relating to inequality of access for people from black and minority ethnic
groups (see Chapter 6).**2009**
Following the publication of Psychosis and Schizophrenia in Children and Young People
(NCCMH, 2013 [full guideline]; NICE, 2013a), for the 2014 guideline the GDG took the
view that the recommendations should be consistent where appropriate. Therefore
the GDG saw the value in advising practitioners of the equivocal evidence regarding
psychological interventions when compared with antipsychotic medication and
recommended that if person wished to try a psychological intervention alone, this
could be trialled over the course of a month or less. Following Psychosis and
Schizophrenia in Children and Young People the GDG also wished to make it explicit
that the options for first episode psychosis should be oral antipsychotic medication
combined with psychological interventions (family intervention and individual
CBT).
Treatment options for first episode psychosis
conjunction with
9.7.10.2 If the person wishes to try psychological interventions (family intervention

and individual CBT) alone without antipsychotic medication, advise that
psychological interventions are more effective when delivered in
conjunction with antipsychotic medication. If the person still wishes to try
psychological interventions alone, then offer family intervention and CBT.
Agree a time (1 month or less) for reviewing treatment options, including
introducing antipsychotic medication. Continue to monitor symptoms, level
of distress, impairment and level of functioning, (including education,
training and employment), regularly. [new 2014]

How to deliver psychological interventions
9.7.10.3 Family intervention should:
• include the person with psychosis or schizophrenia if practical
• be carried out for between 3 months and 1 year
• include at least 10 planned sessions
• take account of the whole family's preference for either single-
family intervention or multi-family group intervention
• take account of the relationship between the main carer and the
person with psychosis or schizophrenia
• have a specific supportive, educational or treatment function and
include negotiated problem solving or crisis management work.
[2009]

• oral antipsychotic medication (see recommendations 10.11.1.2- 10.11.1.13) in
conjunction with
9.7.10.5 Offer family intervention to all families of people with psychosis or

schizophrenia who live with or are in close contact with the service user
(delivered as described in recommendation 9.7.10.3). This can be started
either during the acute phase or later, including in inpatient settings. [2009]
Promoting recovery
9.7.10.6 Offer family intervention to families of people with psychosis or
schizophrenia who live with or are in close contact with the service user.
Deliver family intervention as described in recommendation 9.7.10.3.[2009]
9.7.10.7 Family intervention may be particularly useful for families of people with
psychosis or schizophrenia who have:
• recently relapsed or are at risk of relapse
• persisting symptoms. [2009]

9.7.11.1 For people with schizophrenia from black and minority ethnic groups living
in the UK, does ethnically adapted family intervention for schizophrenia
(adapted in consultation with black and minority ethnic groups to better suit
different cultural and ethnic needs) enable more people in black and
minority ethnic groups to engage with this therapy, and show concomitant
reductions in patient relapse rates and carer distress? [2009]

9.7.11.2 Research is needed to identify the competencies required to deliver effective
family intervention to people with schizophrenia and their carers. [2009]
9.8 PSYCHODYNAMIC AND PSYCHOANALYTICAL

THERAPIES
9.8.1 Introduction
**2009** Psychoanalysis and its derivatives, often termed psychoanalytic and
psychodynamic psychotherapies, originate from the work of Freud in the first
quarter of the 20th century. These approaches assume that humans have an
unconscious mind where feelings that are too painful to face are often held. A
number of psychological processes known as defences are used to keep these
feelings out of everyday consciousness. Psychoanalysis and psychodynamic
psychotherapy aim to bring unconscious mental material and processes into full
consciousness so that the individual can gain more control over his or her life. These
approaches were originally regarded as unsuitable for the treatment of the
psychoses (Freud, 1964). However, a number of psychoanalysts have treated people
with schizophrenia and other psychoses using more or less modified versions of
psychoanalysis (Fromm-Reichmann, 1950; Stack-Sullivan, 1974). Psychoanalytically-
informed approaches to psychotherapy continue to be accessed by people with
schizophrenia today, though the actual psychoanalytic technique is rarely used
(Alanen, 1997). Approaches tend to be modified to favour relative openness on the
part of the therapist, flexibility in terms of content and mode of sessions, holding off
from making interpretations until the therapeutic alliance is solid, and building a
relationship based on genuineness and warmth while maintaining optimal distance
(Gabbard, 1994).
RCTs were undertaken in the 1970s and 1980s to investigate the use of
psychoanalytically-orientated psychotherapy. Research into the effects of psycho-
analytic approaches in the treatment of schizophrenia has been repeated more
recently, with mixed results (Fenton & McGlashan, 1995; Jones et al., 1998; Mari &
Streiner, 2000), leading to the publication of a Cochrane Review on the subject
(Malmberg et al., 2001).
Definition
Psychodynamic interventions were defined as having:
• regular therapy sessions based on a psychodynamic or psychoanalytic
model; and
• sessions that could rely on a variety of strategies (including explorative
insight- orientated, supportive or directive activity), applied flexibly.
To be considered as well-defined psychodynamic psychotherapy, the intervention
needed to include working with transference and unconscious processes.
Psychoanalytic interventions were defined as having:

• regular individual sessions planned to continue for at least 1 year; and

•analysts required to adhere to a strict definition of psychoanalytic
technique.
To be considered as well-defined psychoanalysis, the intervention needed to
involve working with the unconscious and early child/adult relationships.

eligibility criteria used for this section of the guideline, can be found in
Table 81. The primary clinical questions can be found in Box 1. A new systematic
search for relevant RCTs, published since the 2002 guideline, was conducted for the
2009 guideline (further information about the search strategy can be found in
Appendix 20).

In the 2002 guideline, three RCTs (N = 492) of psychodynamic and psychoanalytic
therapies were included. The search for the 2009 guideline identified one new trial.
In total, four RCTs (N = 558) met the inclusion criteria for the 2009 guideline. All of
the trials were published in peer-reviewed journals between 1972 and 2003. In
addition, one study identified in the search for the 2009 guideline was excluded from
the analysis because of an inadequate method of randomisation (further information
about both included and excluded studies can be found in Appendix 22c).
Table 81: Clinical review protocol for the review of psychodynamic and
psychoanalytic therapies
Electronic databases Databases: CINAHL, CENTRAL, EMBASE, MEDLINE, PsycINFO

Studydesign RCT (≥10 participants per arm)

disorders)
Excluded populations Very late onset schizophrenia (onset after age 60) Other psychotic disorders,
such as bipolar disorder, mania or depressive psychosis
Interventions Psychodynamic and psychoanalytic therapies
Mental state (total symptoms, depression) Psychosocial functioning
Quality of life
Adverse events

9.8.4 Psychodynamic and psychoanalytic therapies versus control
For the 2009 guideline review, two RCTs of psychodynamic and psychoanalytic
therapies versus any type of control were included in the meta-analysis.
Additionally, two trials included in the 2002 guideline directly compared the format
of the intervention; one trial compared insight-orientated with reality-adaptive
therapy and another trial compared individual with group therapy 29 (see Table 82
for a summary of the study characteristics). Forest plots and/or data tables for each
outcome can be found in Appendix 23d.

Only one new RCT was identified for the 2009 guideline review (DURHAM2003),
which used a psychodynamic-based intervention as a comparator for CBT. The new
study did not provide any evidence for the effectiveness of psychodynamic
approaches in terms of symptoms, functioning or quality of life.

In the 2002 guideline, the GDG found no clear evidence to support the use of
psychodynamic and psychoanalytic therapies as discrete interventions. The limited
evidence found for the 2009 guideline does not justify changing this conclusion.
However the GDG did acknowledge the use of psychoanalytic and psychodynamic
principles to help healthcare professionals understand the experience of people with
schizophrenia and their interpersonal relationships, including the therapeutic
relationship. Furthermore, the GDG noted that the majority of trials included in the
review assessed the efficacy of classic forms of psychodynamic and psychoanalytic
therapy. However, these approaches have evolved in recent years, partly in response
to a lack of demonstrable efficacy when compared with other interventions in
research trials. At present, the GDG are not aware of any well-conducted RCTs
assessing the efficacy of newer forms of psychodynamic and psychoanalytic therapy.
It is therefore the view of the GDG that further well-conducted research is
warranted.
29Existing subgroups comparing psychodynamic and psychoanalytic therapies with standard care and other
active treatments and psychodynamic therapy with group psychodynamic therapy were also updated. However,
there was insufficient data to draw any conclusions based on these subgroups. Please refer to Appendix 23d for
the forest plots and/or data tables for all subgroup comparisons conducted

Table 82: Summary of study characteristics for psychodynamic and psychoanalytic therapies
Psycho dynamic and Insight-orientated therapy Individual therapy versus

psychoanalytic therapies versus reality adaptive group therapy
versus any control therapy
K (total N) 2 (294) 1 (164) 1 (100)
Study ID DURHAM2003 Gunderson1984 O’Brien1972
May1976
Diagnosis 100% schizophrenia or other 100% schizophrenia or other 100% schizophrenia
related diagnoses (DSM or ICD- related diagnoses (DSM II or III) Or other related diagnoses
10) (DSMII or III)
Baseline severity BPRS: Not reported Not reported
Mean (SD) ~96 (17)
DURHAM2003
Length of treatment Range: 36–104weeks Up to 2 years 20 months
Length of follow-up Up to 3 months:
DURHAM2003
Up to 5 years:
May1976
Setting Inpatient: Inpatient:
May1976 Gunderson1984a
Inpatient and outpatient: Outpatient:

DURHAM2003 O’Brien1972b
Note. aTreatment was initiated in the inpatient setting and continued in a community setting up on discharge.
bAll participants were newly discharge

9.8.7.1 Healthcare professionals may consider using psychoanalytic and
psychodynamic principles to help them understand the experiences of
people with psychosis or schizophrenia and their interpersonal
relationships. [2009]

9.8.8.1 A pilot RCT should be conducted to assess the efficacy of contemporary
forms of psychodynamic therapy when compared with standard care and
other active psychological and psychosocial interventions. [2009]
9.9 PSYCHOEDUCATION
9.9.1 Introduction
Psychoeducation, in its literal definition, implies provision of information and
education to a service user with a severe and enduring mental illness, including
schizophrenia, about the diagnosis, its treatment, appropriate resources, prognosis,
common coping strategies and rights (Pekkala & Merinder, 2002).
In his recent review of the NHS, Darzi (2008) emphasised the importance of
‘empowering patients with better information to enable a different quality of
conversation between professionals and patients’. Precisely what and how much
information a person requires, and the degree to which the information provided is
understood, remembered or acted upon, will vary from person to person.
Frequently, information giving has to be ongoing. As a result, psychoeducation has
now been developed as an aspect of treatment in schizophrenia with a variety of
goals over and above the provision of accurate information. Some psychoeducation
involves quite lengthy treatment and runs into management strategies, coping
techniques and role-playing skills. It is commonly offered in a group format. The
diversity of content and information covered, as well as the formats of delivery, vary
considerably, so that psychoeducation as a discrete treatment can overlap with
family intervention, especially when families and carers are involved in both.
Desired outcomes in studies have included improvements in insight, treatment
adherence, symptoms, relapse rates, and family knowledge and understanding
(Pekkala & Merinder, 2002).
Definition
Psychoeducational interventions were defined as:
• any programme involving interaction between an information
provider and service users or their carers, which has the primary aim
of offering information about the condition; and
• the provision of support and management strategies to service users
and carers.

To be considered as well defined, the educational strategy should be tailored to the
need of individuals or carers.

relevant RCTs, published since the 2002 guideline, was conducted for the 2009
20).
Table 83: Clinical review protocol for the review of psychoeducation
Electronic databases Databases: CINAHL, CENTRAL, EMBASE, MEDLINE, PsycINFO

Study design RCT (≥10 participants per arm and ≥6 weeks’ duration)
Patient population Adults (18+) with schizophrenia (schizophrenia-related disorders)

Excluded populations Very late onset schizophrenia (onset after age60) Other psychotic
People with coexisting learning difficulties, significant physical or
sensory difficulties, or substance misuse
Interventions Psychoeducation
Quality of life
Ad
In the 2002 guideline, ten RCTs (N = 1,070) of psychoeducation were included. The
search for the 2009 guideline identified three papers providing follow-up data to
existing trials and ten new trials. In the 2002 guideline, one study (Posner1992)
included in the family intervention review was reclassified as psychoeducation for
the 2009 guideline. In total, 21 trials (N = 2,016) met the inclusion criteria for the 2009
guideline review. All were published in peer-reviewed journals between 1987 and
2008 (further information about both included and excluded studies can be found in
Appendix 22c).
9.9.4 Psychoeducation versus control

For the 2009 guideline, four of the included studies (Jones2001; SIBITZ2007;
Smith1987; XIANG2007) only included a direct comparison of different types of
psychoeducation and one trial (AGARA2007) did not provide any useable data, so
16 trials of psychoeducation versus any type of control were included in the meta-

analysis (see Table 84 for a summary of the study characteristics). Subgroup analyses
were used to examine the impact of the type of comparator (eight trials used
standard care as the comparator and eight trials used another active treatment 30).
Forest plots and/or data tables for each outcome can be found in Appendix 23d.

There is no new robust evidence for the effectiveness of psychoeducation on any of
the critical outcomes. In particular, there are no new UK-based RCTs meeting the
GDG’s definition of psychoeducation.

In the 2002 guideline, the GDG found it difficult to distinguish psychoeducation
from the provision of good-quality information as required in standard care, and
from good-quality family engagement, where information is provided with family
members also present. There is clearly an overlap between good standard care and
psychoeducation, and between psychoeducation and family intervention. It is
noteworthy that most of the studies reviewed did not take place in the UK, and the
nature and quality of the information provision in standard care may differ from
services in the UK setting. The evidence found for the 2009 guideline does not justify
making a recommendation.
30Existing subgroup comparisons exploring the country of the trial, format of the intervention, number of
treatment sessions, duration of treatment and patient characteristics were also updated. However, there was
insufficient data to draw any conclusions based on these subgroups. Please refer to Appendix 23d for the forest
plots and/or data tables for all subgroup comparisons conducted.

Table 84: Summary of study characteristics for psychoeducation
Psychoeducation versus any Psychoeducation versus Psychoeducation versus other

control standard care active treatments
K (total N) 16 (1610) 8 (966) 8 (644)
Study ID Atkinson1996 Atkinson1996 BECHDOLF2004
Bauml1996 Bauml1996 CATHER2005
BECHDOLF2004 CHABANNES2008 CHAN2007A Hornung1995a
CATHER2005 CunninghamOwnes2001 Lecompte1996
CHABANNES2008 Hayashi2001 Merinder1999
CHAN2007A Macpherson1996 SHIN2002
CunninghamOwens2001 Posner1992 XIANG2006
Hayashi2001 VREELAND2006
Hornung1995a
Lecompte1996
Macpherson1996
Merinder1999
Posner1992
SHIN2002
VREELAND2006
XIANG2006
Diagnosis 100% schizophrenia or other 100% schizophrenia or other 100% schizophrenia or other
related diagnoses (DSM or ICD-10) related diagnoses (DSM or ICD-10) related diagnoses (DSM or ICD-
10)

Psychoeducation versus any Psychoeducation versus Psychoeducation versus other

control standard care active treatments
Baseline severity BPRS total: Not reported BPRS total:
Mean (SD) range: Mean (SD) range:
~29 (7) to ~92 (8) ~29 (7) to ~92 (8)
PANSS total: PANSS total:

Mean (SD) range: Mean (SD) range:
~14 (5) to ~51 (13) ~14 (5) to ~51 (13)
Length of treatment Range: 2– 52 weeks Range: 4– 52 weeks Range: 2–16 weeks
Length of follow-up Range: 3–60months Range: 3–24months Range: 12–60months

Setting Inpatient: Inpatient: Inpatient:
BECHDOLF2004 CunninghamOwens2001b BECHDOLF2004
CHAN2007A Hayashi2001 CHAN2007A
CunninghamOwens2001b VREELAND2006
Hayashi2001
VREELAND2006
Outpatient: Outpatient: Outpatient:
Atkinson1996 Atkinson1996 CATHER2005
Bauml1996 Bauml1996 Hornung1955a
CATHER2005 Macpherson1996 Merinder1999
Hornung1995a Posner1992 SHIN2002
Macpherson1996 XIANG2006
Merinder1999
Posner1992
SHIN2002
XIANG2006
Inpatient and outpatient: Inpatient and outpatient:

CHABANNES2008 CHABANNES2008
Note. aMulti-modal intervention.
b Participants were recruited as inpatients prior to discharge.

9.10 SOCIAL SKILLS TRAINING
9.10.1 Introduction
An early psychological approach to the treatment of schizophrenia involved the
application of behavioural theory and methods with the aim of normalising
behaviour (Ayllon & Azrin, 1965), improving communication or modifying speech
(Lindsley, 1963). Given the complex and often debilitating behavioural and social
effects of schizophrenia, social skills training was developed as a more sophisticated
treatment strategy derived from behavioural and social learning traditions (see
Wallace and colleagues (1980) for a review). It was designed to help people with
schizophrenia regain their social skills and confidence, improve their ability to cope
in social situations, reduce social distress, improve their quality of life and, where
possible, to aid symptom reduction and relapse prevention.
Social skills training programmes begin with a detailed assessment and behavioural
analysis of individual social skills, followed by individual and/or group
interventions using positive reinforcement, goal setting, modelling and shaping.
Initially, smaller social tasks (such as responses to non-verbal social cues) are
worked on, and gradually new behaviours are built up into more complex social
skills, such as conducting a meaningful conversation. There is a strong emphasis on
homework assignments intended to help generalise newly learned behaviour away
from the treatment setting.
Although this psychosocial treatment approach became very popular in the US and
has remained so (for example, (Bellack, 2004)) since the 1980s it has had much less
support in the UK, at least in part as a result of doubts in the UK about the evidence
of the capacity of social skills training to generalise from the treatment situation to
real social settings (Hersen & Bellack, 1976; Shepherd, 1978). No new studies,
therefore, have been conducted of social skills training in the UK. Instead, the
evidence base is largely derived from North America and, increasingly, from China
and Southeast Asia.
Definition
Social skills training was defined as:
• a structured psychosocial intervention (group or individual) that aims
to:
- enhance social performance, and
- reduce distress and difficulty in social situations.
The intervention must:
• include behaviourally-based assessments of a range of social and
interpersonal skills, and
• place importance on both verbal and non-verbal communication, the
individual’s ability to perceive and process relevant social cues, and
respond to and provide appropriate social reinforcement.

A new systematic search for relevant RCTs published since the 2002 guideline was
conducted for the 2009 guideline. Information about the databases
searched and the eligibility criteria used for this section of the guideline can be
found in Table 85 (further information about the search strategy can be found in
Appendix 20).

In the 2002 guideline, nine RCTs (N = 436) of social skills training were included.
One RCT from the 2002 guideline (Finch1977) was removed from the 2009 guideline
analysis because of inadequate numbers of participants, and one RCT
(Eckmann1992) was reclassified as social skills training and included in the analysis.
The search for the 2009 guideline identified 14 new trials. In total, 23 trials (N =
1,471) met the inclusion criteria for the 2009 guideline. All were published in peer-
reviewed journals between 1983 and 2007 (further information about both included
and excluded studies can be found in Appendix 22c).
Table 85: Clinical review protocol for the review of social skills training

PsycINFO
Study design RCT (≥10 participants per arm and ≥6 weeks’
duration)
Patient population Adults (18+) with schizophrenia

(including schizophrenia-related disorders)
Excluded populations Very late onset schizophrenia (onset after age 60)
Other psychotic disorders, such as bipolar disorder,
mania or depressive psychosis
Interventions Social skills training

Quality of life
Adverse events

9.10.4 Social skills training versus control
For the 2009 guideline review, one of the included studies (GLYNN2002) only
included a direct comparison of different types of social skills and two trials
(GUTRIDE1973, KERN2005) did not provide any useable data for any of the critical
outcomes listed in the review protocol. Thus, in total 20 trials of social skills training
versus any type of control were included in the meta-analysis (see Table 86 for a
summary of the study characteristics). Subgroup analyses were used to examine the
impact of the type of comparator 31 (ten trials used standard care as the comparator
and ten trials used another active treatment). Forest plots and/or data tables for each
outcome can be found in Appendix 23d.

The review found no evidence to suggest that social skills training is effective in
improving the critical outcomes. None of the new RCTs were UK based, with most
new studies reporting non-significant findings. There was limited evidence for the
effectiveness of social skills training on negative symptoms. However this evidence
is primarily drawn from non-UK studies and is largely driven by one small study
(RONCONE2004) that contains multiple methodological problems.

In the 2002 guideline, the GDG found no clear evidence that social skills training was
effective as a discrete intervention in improving outcomes in schizophrenia when
compared with generic social and group activities, and suggested that the evidence
shows little if any consistent advantage over standard care. It is noteworthy that
although a review published since the 2002 guideline (Kurtz & Mueser, 2008)
indicated effects for social functioning, symptom severity and relapse, this may be
attributed to the inclusion of a number of studies that are beyond the scope of the
current definition of social skills used in the present review. In particular, a number
of papers were included that assessed vocational and supported employment-based
interventions. Consequently, the evidence found for the 2009 guideline does not
justify changing the conclusions drawn in the 2002 guideline.
9.10.7.1 Do not routinely offer social skills training (as a specific intervention) to
people with psychosis or schizophrenia. [2009]**2009**
31Existing
subgroup comparisons exploring the duration of treatment and treatment setting were also updated.
However, there was insufficient data to draw any conclusions based on these subgroups. Please refer to
Appendix 23d for the forest plots and/or data tables for all subgroup comparisons conducted

Table 86: Summary of study characteristics for social skills training
Social skills training versus Social skill straining versus Social skills training versus
any control standard care other active treatments
K (total N) 20 (1215) 10 (541) 10 (674)
Study ID Bellack1994 Bellack1984 BROWN1983
BROWN1983 CHIEN2003 Dobson1995
CHIEN2003 CHOI2006 Eckmann1992
CHOI2006 Daniels1998 Hayes1995
Daniels1998 GRANHOLM2005a Liberman1998
Dobson1995 PATTERSON2003 Lukoff1986
Eckmann1992 Peniston1988 Marder1996
GRANHOLM2005a RONCONE2004 NG2007
Hayes1995 UCOK2006 PATTERSON2006
Liberman1998 VALENCIA2007a PINTO1999a
Lukoff1986a
Marder1996
NG2007
PATTERSON2003
PATTERSON2006
PINTO1999a
Peniston1988
RONCONE2004
UCOK2006
VALENCIA2007a

Social skills training versus Social skills training versus Social skills training versus
any control standard care other active treatments
Diagnosis 100% schizophrenia or other 100% schizophrenia or other 100% schizophrenia or other
related diagnoses (DSM or ICD-10) related diagnoses (DSM or ICD-10) related diagnoses (DSM or ICD-10)

Mean (SD) ~47 (10) Mean (SD) ~ 41 (7) Mean (SD) ~47 (10)
Hayes1995 UCOK2006 Hayes1995
Mean (SD) ~40 (10) Mean (SD) ~40 (10)
NG2007 NG2007
Mean (SD) ~82 (21) Mean (SD) ~82 (21)
PINTO1999a PINTO1999a
Mean (SD) ~41 (7)
UCOK2006

Mean (SD) ~54 (14) Mean (SD) ~54 (14) Mean (SD) ~61 (3)
GRANHOLM2005a GRANHOLM2005a PATTERSON2006
Mean (SD) ~61 (3) Mean (SD) ~ 112 (27)
PATTERSON2006 VALENCIA2007a

Length of treatment Range: 4–104 weeks Range: 4–52 weeks Range: 8–104 weeks
Length of follow-up Up to 12 months: Up to 12 months: Up to 12 months:

Bellack1984 Bellack1984 Hayes1995
CHIEN2003 CHIEN2003 PATTERSON2006
Hayes1995 PATTERSON2003
PATTERSON2003
PATTERSON2006
Up to 24 months: Up to 24 months:
Liberman1998 Liberman1998
Lukoff1986 Lukoff1986
Setting Inpatient: Inpatient: Inpatient:

BROWN1983 CHIEN2003 BROWN1983
CHIEN2003 Peniston1988 Luckoff1986
Lukoff1986 RONCONE2004 NG2007
NG2007
Peniston1988
RONCONE2004
Outpatient:
Outpatient: Outpatient:
CHOI2006
CHOI2006 Liberman1998
GRANHOLM2005a
GRANHOLM2005a Marder1996
Liberman1998
UCOK2006

Marder1996 VALENCIA2007a
UCOK2006
VALENCIA2007a
Inpatient and outpatient:

Inpatient and outpatient:
Daniels1998 Inpatient and outpatient:
Daniels1998 Eckmann1992
Eckmann1992
Hayes1995 Hayes1995
PINTO1999a PINTO1999a
Otherb: Otherb: Otherb:

Bellack1984 Bellack1984 Dobson1995
Dobson1995 PATTERSON2003 PATTERSON2006
PATTERSON2003
Note. Multi-modal interventions.SO 2006
a
bOther settings include board and care facilities, and day hospitals

9.11 PSYCHOLOGICAL MANAGEMENT OF TRAUMA IN
PSYCHOSIS AND SCHIZOPHRENIA
9.11.1 Introduction
There has been a growing interest in the relationship between psychosis (including
schizophrenia) and trauma over the last decade. Studies of individuals who have
experienced psychosis and schizophrenia have found that between 50 and 98%
report having being exposed to at least one traumatic event in their lives (Read et al.,
2005).
A recent review discussing childhood adversity and mental health problems

suggests that factors related to the mother (for example, high levels of stress during
pregnancy, poor nutrition, and mother’s ill health), as well as childhood adversity,
can have a negative impact on an individual’s future mental health (Read & Bentall,
2012). Investigating early adversity, Morgan et al (2007) found that loss of a parent
through separation or death in young people under the age of 16 years was
associated with an increased risk of psychosis. A review by Read et al (2005)
demonstrated there was a strong relationship between those people who had
experienced physical and sexual abuse as children and the presence of symptoms of
schizophrenia. In a Dutch prospective study, Janssen et al (2004) controlled for a
number of potential variables including substance misuse and a family history of
psychosis, and found that those who had been subjected to any form of childhood
abuse were over seven times more likely to experience psychosis. A number of
studies have found a ‘dose response’, with more severe or enduring abuse increasing
the risk of developing psychosis. This was clearly illustrated in a study by Shevlin et
al (2008) that found that the likelihood of developing psychosis increased as the
number of traumatic experiences to which an individual had been exposed also
increased. Those who had experienced five or more types of trauma were 198 times
more likely to have a diagnosis of psychosis than those who had not experienced any
adversity.
Varese et al (2012) examined the relationship between psychosis and childhood

adversity (physical, sexual and emotional abuse, neglect, bullying and parental
death or separation) by conducting a meta-analysis that included 36 studies (n =
79,397). A significant association was found between the two, with an odds ratio of
2.78. Based on their findings the authors stated that if these particular forms of
childhood adversity were eliminated, cases of psychosis would be reduced by a
third. The authors also investigated the severity of the trauma and its relationship
with psychosis. Nine out of ten of the studies that had researched a so-called 'dose
effect' had found this, revealing that the likelihood of psychosis increases the more
severe or prolonged the exposure to adversity. Trauma within this population is not
restricted to childhood: incidence of assaults in adulthood are also elevated: up to
59% of individuals report sexual assault and up to 87% report physical assault
(Grubaugh et al., 2011).

Not all adversity, however intolerable the subjective experience, fulfils diagnostic
criteria to be classed as a ‘trauma’. The objective definition of what does and does
not constitute a trauma evidently impacts on what symptoms can be classified as
part of a genuine post-traumatic stress disorder. Despite this, the prevalence of PTSD
in those diagnosed with a psychotic disorder ranges from 12 to 29% (Achim et al.,
2011; Buckley et al., 2009), which is a much higher rate than in the general
population where prevalence is estimated to be between 0.4 and 3.5% (Alonso et al.,
2004; Creamer et al., 2001; Darves-Bornoz et al., 2008). It has been suggested that
there are similarities in vulnerability to PTSD and schizophrenia as a result of the
cognitive processing of traumatic events, and the way in which information is
processed and stored (Steel, 2011).
One issue that is commonly raised is that of the reliability of disclosures of

childhood abuse among those with psychosis. Studies investigating this found
ccorroborating evidence for reports of childhood sexual abuse by psychiatric
patients in 74% (Herman & Schatzow, 1987) and 82% (Read et al., 2003). One study
that focused specifically on the reports of those with a diagnosis of schizophrenia,
found that the problem of false allegations of sexual assault was no different than in
the general population (Darves-Bornoz et al., 1995).
Current practice
Though not all of those presenting with psychosis or schizophrenia will have been
exposed to early adversity, the significance of the relationship between them means
there is a high likelihood that there will be a history of trauma. Currently, however,
the question of what constitutes appropriate help for those with psychosis and
schizophrenia with a history of trauma is unclear. NICE guidance recommends
trauma-focused CBT (including prolonged exposure) and eye movement
desensitisation and reprocessing (EMDR) as safe and effective interventions for
those with PTSD. Unfortunately because people with psychotic disorders are often
excluded from PTSD research trials, there is insufficient evidence to demonstrate
whether these particular interventions are equally safe and effective in this
population.
Nevertheless, service users presenting with psychosis and schizophrenia who have
trauma histories have not been excluded from trials testing the efficacy of CBT for
psychotic disorders. Moreover, no adverse effects or differences in outcomes have
been reported for this particular group within these trials.
Definition and aim of intervention

The aim of this review was to evaluate the effectiveness and safety of psychological
interventions for trauma in a population of people with psychosis and
schizophrenia.
Psychological interventions were included if they aimed to reduce PTSD symptoms

or other related distress. PTSD symptoms could be a result of life events, a reaction

to psychosis symptoms, or trauma as a result of experiencing a first episode
psychosis.
9.11.2 Clinical review protocol (psychological management of trauma)

guideline, can be found in Table 87 (a complete list of review questions and
protocols can be found in Appendix 6; further information about the search strategy
can be found in Appendix 13.
Table 87: Clinical review protocol for the review of psychological management
of trauma
Review question For adults with psychosis and schizophrenia, what are the benefits
and/or potential harms of psychological management strategies for
previous trauma compared to treatment as usual or another
intervention?
Objectives To evaluate the clinical effectiveness of psychological interventions for
trauma for people with psychosis and schizophrenia.
Population Included
Adults (18+) with schizophrenia (including schizophrenia-related
disorders such as schizoaffective disorder and delusional disorder) or
psychosis.
Intervention(s) Psychological interventions for trauma
Critical outcomes • Anxiety symptoms (including PTSD)
• Depression symptoms
o Total symptoms
o Positive symptoms
o Negative symptoms
• Response / Relapse
o Relapse (as defined in study)
o Response (improvement in symptoms)
• Dropout (proxy measure for acceptability)
o Withdrawal due to adverse event
o Loss to follow-up, any reason
Electronic databases Core: CDSR, CENTRAL, DARE, Embase, HTA, MEDLINE,
PreMEDLINE
Date searched • RCT: database inception to June 2013
• SR: 1995 to June 2013

Analyses were conducted for follow-up using data from the last
follow-up point reported within the time-point groupings
Sub-analysis
Where data were available, sub-analyses were conducted of studies
with >75% of the sample described as having a primary diagnosis of
Where data were available, sub-analyses were conducted for

UK/Europe studies.
9.11.3 Studies considered 32

One RCT (N = 66) met the eligibility criteria for this review: JACKSON2009 (Jackson
et al., 2009). Further information about the included and excluded studies can be
found in Appendix 15a.
The single included trial had sufficient data to be included in the statistical analysis.
This trial involved a comparison between cognitive therapy-based recovery
intervention (CRI) plus treatment as usual (case management and antipsychotic
medication) compared with treatment as usual alone for the treatment of first
episode psychosis-related trauma. Table 88 provides an overview of the included
trial.
Table 88: Study information table for trials comparing psychological trauma
interventions with any alternative management strategy
Psychological management of trauma versus any

Total no. of trials (k); participants (N) k = 1; (N = 66)
Study ID JACKSON2009
Country UK
Year of publication 2009
Mean Age of participants 23.3 years
Mean percentage of participants with 100%
primary diagnosis of psychosis or
Mean gender % women 25.7%
Length of treatment 26 weeks
Length of follow-up 6 months
JACKSON2009
Intervention type Cognitive therapy-based recovery intervention (CRI)
plus TAU (k = 1)
Comparisons Case management and antipsychotic medication (k =
1)

9.11.4 Clinical evidence for psychological management of trauma
Table 89: Summary of findings table for cognitive therapy-based recovery

intervention compared with treatment as usual
Patient or population: Adults with psychosis and schizophrenia with trauma

Intervention: Cognitive therapy + TAU
Comparison: TAU
TAU Cognitive therapy + TAU CI) (GRADE)
Anxiety symptoms - end N/A The mean anxiety symptoms, end of intervention in the N/A 46 ⊕⊕⊝⊝
of intervention intervention groups was (1 study) Low1,2
0.34 standard deviations lower (0.93 lower to 0.24 higher)
Anxiety symptoms - up N/A The mean anxiety symptoms, up to 6 months’ follow-up in the N/A 46 ⊕⊕⊝⊝
to 6 months’ follow-up intervention groups was 0.47 standard deviations lower (1 study) Low1,2
Depression symptoms - N/A The mean depression symptoms, end of intervention in the N/A 46 ⊕⊕⊝⊝
end of intervention intervention groups was (1 study) Low1,2
0.29 standard deviations lower (0.87 lower to 0.3 higher)
Depression symptoms - N/A The mean depression symptoms, up to 6 months’ follow-up in N/A 46 ⊕⊕⊝⊝
up to 6 months’ follow- the intervention groups was 0.05 standard deviations lower (1 study) Low1,2
up (0.63 lower to 0.52 higher)
Missing data, any reason RR 1.94 66 ⊕⊕⊝⊝
Study population
- end of intervention (0.85 to (1 study) Low1,2
200 per 388 per 1000 4.43)
1000 (170 to 886)
200 per 388 per 1000
1000 (170 to 886)
Missing data, any reason RR 1.94 66 ⊕⊕⊝⊝
Study population
- up to 6 months’ follow- (0.85 to (1 study) Low1,2
up 200 per 388 per 1000 4.43)
1000 (170 to 886)
200 per 388 per 1000
1000 (170 to 886)
Note. CI = confidence interval; RR = risk ratio; TAU = treatment as usual
*The basis for the assumed risk (for example, the median control group risk across studies) is provided in the footnotes. The
corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the
intervention (and its 95% CI).
1 Studies included at moderate risk of bias.
Low quality evidence from one study with 46 participants showed no significant
difference between CRI and TAU in anxiety or depression symptoms at the end of
the intervention or at 6 months’ follow-up. There was no statistically significant
difference between CRI and TAU in the number of participants who dropped out of
the study although a trend showing fewer dropouts in the TAU arm was observed.
No data were available for the critical outcomes of psychosis symptoms, or relapse
and response rates.

Overall there is inconclusive evidence concerning the efficacy of the psychological
management of trauma and a specific cognitive therapy-based recovery intervention
for the treatment of trauma in people with first episode psychosis. In addition,
although this review found no statistically significant difference between the active
intervention and control in dropouts from the intervention, a trend favouring the
control arm was observed suggesting that the intervention may not have been well
tolerated. However, due to the limited evidence, and lack of trials evaluating other
interventions in this population, no firm conclusions can be drawn.

No studies assessing the cost effectiveness of psychological interventions for trauma
in adults with psychosis and schizophrenia were identified by the systematic search
of the economic literature undertaken for this guideline. Details on the methods used
for the systematic search of the economic literature are described in Chapter 3.

The GDG decided to focus on the following, which were considered to be critical:
For trauma-focused symptoms:

• Anxiety symptoms (including PTSD)
• Depression symptoms
To evaluate if psychological intervention for trauma was contraindicated in a

population of people with psychosis and schizophrenia:
• Symptoms of psychosis (total, positive, negative)
• Response/relapse
To evaluate the acceptability of the intervention:

• Dropout (for any reason)
Trade-off between clinical benefits and harms:

In people with psychosis and schizophrenia who are experiencing trauma-related
symptoms, the GDG considered that it was important to assess the potential harms
of psychological interventions for trauma. The GDG judged that the evidence did
not show any benefit of psychological interventions for trauma in this population
but importantly did not observe any indication of harm. However, the latter was as a
result of a lack of data and thus there is still come uncertainty about the effects of
these interventions on symptoms of psychosis and schizophrenia.
Trade-off between net health benefits and resource use:

There were no health economic studies that attempted to assess the cost effectiveness
associated with psychological interventions for trauma in a population of people

with psychosis and schizophrenia. Due to the lack of clinical data pertaining to the
response and relapse rates, and effects of these interventions on symptoms of
psychosis and schizophrenia, it was decided that formal economic modelling of such
interventions in this area would not be useful in decision-making. The study
included in clinical review point to a resource use that is more intensive than usual
care (that is, intervention was provided in addition to usual care), which implies that
such psychological interventions for trauma in a population of people with
psychosis and schizophrenia is likely to be more costly than usual care. However,
this does not exclude the possibility of such interventions being cost effective when
compared to usual care since the clinical evidence is inconclusive and even small
differences in effects and costs could potentially result in a cost-effective
intervention.

The quality of the evidence was low. The two reasons for downgrading the evidence
were: (1) potential risk of bias in the single included trial and (2) moderate
imprecision in the results. The available evidence was directly applicable to the
population of interest but the inclusion of only a single trial meant that the GDG
could not consider issues around inconsistency. The GDG thought that there was a
lack of published research in this topic area and thus could not be certain of the
presence of publication bias.
The GDG felt that it was of crucial importance that symptoms of trauma are
identified and assessed in first episode psychosis in order to identify those who may
be experiencing intrusions as a result of first episode psychosis and this should be
reflected in recommendations. The GDG discussed the need for improved access to
PTSD services for people with psychosis and schizophrenia. The GDG felt this was
especially important for those experiencing first episode psychosis. The GDG
thought that as there was no evidence that a psychological intervention for trauma
was contraindicated in people experiencing first episode psychosis therefore
recommendations in the PTSD guideline were applicable to people with psychosis
and schizophrenia.
9.11.8.1 Assess for post-traumatic stress disorder and other reactions to trauma
because people with psychosis or schizophrenia are likely to have
experienced previous adverse events or trauma associated with the
development of the psychosis or as a result of the psychosis itself. For people
who show signs of post-traumatic stress, follow the recommendations in
Post-traumatic stress disorder (NICE clinical guideline 26). [new 2014]

9.12 **2009**RECOMMENDATIONS (ACROSS ALL
TREATMENTS) 33
9.12.1 Principles in the provision of psychological therapies
9.12.1.1 When providing psychological interventions, routinely and systematically
monitor a range of outcomes across relevant areas, including service user
satisfaction and, if appropriate, carer satisfaction. [2009]
9.12.1.2 Healthcare teams working with people with psychosis or schizophrenia
should identify a lead healthcare professional within the team whose
responsibility is to monitor and review:
• access to and engagement with psychological interventions
• decisions to offer psychological interventions and equality of access
across different ethnic groups. [2009]
9.12.1.3 Healthcare professionals providing psychological interventions should:
• have an appropriate level of competence in delivering the
intervention to people with psychosis or schizophrenia
• be regularly supervised during psychological therapy by a
competent therapist and supervisor. [2009]
9.12.1.4 Trusts should provide access to training that equips healthcare professionals
with the competencies required to deliver the psychological therapy
interventions recommended in this guideline. [2009]**2009**

9.12.2.1 What is the clinical and cost effectiveness of psychological intervention
alone, compared with treatment as usual, in people with psychosis or
schizophrenia who choose not to take antipsychotic medication?( See
Appendix 10 for further details) [2014]
9.12.2.2 What is the benefit of a CBT-based trauma reprocessing intervention on
PTSD symptoms in people with psychosis and schizophrenia (See Appendix
10 for further details) [2014]
33Recommendations for specific interventions can be found at the end of each review (see the beginning of this
chapter for further information).

10 PHARMACOLOGICAL
INTERVENTIONS IN THE
TREATMENT AND MANAGEMENT
OF SCHIZOPHRENIA
This chapter has been partially updated. Most sections remain unchanged from the
2009 guideline; however some of the recommendations have been updated to bring
them in line with the recommendations from Psychosis and Schizophrenia in Children
and Young People. This was considered necessary to avoid discrepancies between the
child and adult guidelines, particularly regarding early intervention. Consequently
new sections have been added to the evidence to recommendations section. In
addition some recommendations from the 2009 guideline have been amended to
improve the wording and structure with no important changes to the context and
meaning of the recommendation.
Sections of the guideline where the evidence has not been updated since 2002 are
marked as **2002**_**2002** and where the evidence has not be updated since 2009,
marked by asterisks (**2009**_**2009**).
Please note that all references to study IDs in sections that have not been updated in
this chapter can be found in Appendix 22b.
**2009** The term ‘first-generation antipsychotics’ (FGAs) is used to refer to drugs

that in the 2003 guideline were called ‘conventional’ or ‘typical’ antipsychotics.
Likewise, the term ‘second-generation antipsychotics’ (SGAs) is used to refer to
drugs that were called ‘atypical’ antipsychotics in the 2003 guideline. This
terminology is used here because it is widely used in the literature; it should not be
taken to suggest that FGAs and SGAs represent distinct classes of antipsychotics (see
Section 10.4.1 for further discussion of this issue).
For this chapter, there view of evidence is divided into the following areas:
• initial treatment with oral antipsychotic medication (Section 10.2)

• oral antipsychotics in the treatment of the acute episode Section 10.3
• promoting recovery in people with schizophrenia that is in remission –
pharmacological relapse prevention (Section 10.4)
• promoting recovery in people with schizophrenia whose illness has not
responded adequately to treatment (Section 10.5)
• combining antipsychotic medication with another antipsychotic
(Section 10.5.10)
• treatment with depot/long-acting injectable antipsychotic medication
(Section 10.6)

• side effects of antipsychotic medication, focusing on metabolic and
neurologic adverse events—these were considered a priority by the
GDG and were also highlighted as areas of concern by service users
(Section 10.7)
• effectiveness of antipsychotic medication (Section 10.8)
• health economics (Section 10.9).
Because of the nature of the evidence, all recommendations can be found in Section
10.11 at the end of the chapter (rather than after each subsection), preceded by
Section 10.10 (linking evidence to recommendations) that draws together the clinical
and health economic evidence and provides a rationale for the recommendations.
10.1 INTRODUCTION
Antipsychotic drugs have been the mainstay of treatment of schizophrenia since the
1950s. Initially used for the treatment of acute psychotic states, their subsequent use
to prevent relapse led to these drugs being prescribed for long-term maintenance
treatment, either as oral preparations or in the form of long-acting injectable
preparations (‘depots’).
Although a number of different classes of drugs have antipsychotic activity, the

primary pharmacological action of antipsychotic drugs is their antagonistic effect on
the D2 dopamine receptors. Indeed, the potency of a drug’s antipsychotic effect is at
least in part determined by its affinity for the D2 receptor (Agid et al., 2007; Kapur &
Remington, 2001; Snyder et al., 1974), an association that informed the dopamine
hypothesis of schizophrenia. It is worth noting, however, that antipsychotic drugs
are also of use in the treatment of other psychotic disorders, their dopamine-
blocking activity probably again being central to their pharmacological efficacy.
Uses of antipsychotics
In the treatment and management of schizophrenia, antipsychotics are currently
used for the treatment of acute episodes, for relapse prevention, for the emergency
treatment of acute behavioural disturbance (rapid tranquillisation) and for symptom
reduction. They are available as oral, intramuscular (IM) and intravenous (IV)
preparations, or as medium- or long-acting depot IM preparations. In the UK,
clozapine is only licensed for use in people with ‘treatment-resistant’ schizophrenia,
defined by the manufacturers’ Summary of Product Characteristics (SPC) as a ‘lack
of satisfactory clinical improvement despite the use of adequate doses of at least two
different antipsychotic agents, including an atypical antipsychotic agent, prescribed
for adequate duration’.
Antipsychotics are usually prescribed within the recommended SPC dosage range
and there is little evidence to support the use of higher dosage or combination with
another antipsychotic if monotherapy proves to be ineffective (Royal College of
Psychiatrists, 2006; Stahl, 2004). Antipsychotics are also used in combination with a
range of other classes of drugs, such as anticonvulsants, mood stabilisers,

anticholinergics, antidepressants and benzodiazepines. Clinicians may augment
antipsychotics with such drugs for several reasons:
• where there is a lack of effective response to antipsychotics alone
• for behavioural control
• for the treatment of the side effects of antipsychotics
• for the treatment of comorbid or secondary psychiatric problems, such
as depression and anxiety.
Although such augmentation strategies are commonly used in clinical practice, they
are outside the scope of this guideline. It is anticipated that a future guideline will
address the evidence base for these interventions.
Antipsychotic dose
The current British National Formulary (BNF) is the most widely used reference for
the prescription of medicines and the pharmacy industry within the UK, and a
complete SPC for all the drugs referred to in this guideline can be found in the
Electronic Medicines Compendium (http://emc.medicines.org.uk/). The
recommended dose ranges listed in the BNF normally echo the information
contained in the manufacturers’ SPC, as well as advice from an external panel of
experts to ensure that the SPC recommendations on issues such as dose range reflect
current good practice (‘standard dosing’). ‘Standard doses’ are identified as doses that
fall within the range likely to achieve the best balance between therapeutic gain and
dose-related adverse effects. However, with up to a third of people with
schizophrenia showing a poor response to antipsychotic medication, there has been a
tendency for higher doses to be prescribed: surveys of prescribing practice suggest
that doses of antipsychotics exceeding BNF limits, either for a single drug or through
combining antipsychotics, continue to be commonly used (Harrington et al., 2002;
Lehman & Steinwachs, 1998; Paton et al., 2008).
In an attempt to increase the rate or extent of response, ‘loading doses’ and rapid
dose escalation strategies have been employed (Kane & Marder, 1993); studies have
failed to show any advantage for such a strategy in terms of speed or degree of
treatment response (Dixon et al., 1995). The Schizophrenia Patient Outcomes
Research Team (1998) concluded that in the treatment of acute episodes of
schizophrenia ‘massive loading doses of antipsychotic medication, referred to as
“rapid neuroleptization,” should not be used’.
Evidence suggests that drug-naïve patients and those experiencing their first episode
of schizophrenia respond to doses of antipsychotic drugs at the lower end of the
recommended dosage range (Cookson et al., 2002; McEvoy et al., 1991; Oosthuizen et
al., 2001; Remington et al., 1998; Tauscher & Kapur, 2001).
Relapse prevention
For people with established schizophrenia, the chance of relapse while receiving
continuous antipsychotic medication appears to be about a third of that on placebo
(Marder & Wirshing, 2003). Risk factors for relapse of illness include the presence of

persistent symptoms, poor adherence to the treatment regimen, lack of insight and
substance use, all of which can be reasonable targets for intervention.
Stopping antipsychotic medication in people with schizophrenia, especially

abruptly, dramatically increases the risk of relapse in the short to medium term,
although even with gradual cessation about half will relapse in the succeeding 6
months (Viguera et al., 1997). Low-dose prescribing and the use of intermittent
dosing strategies (with medication prompted by the appearance of an individual’s
characteristic early signs of relapse) have also been suggested in the past as ways to
minimise side effects in the long-term. However, when these were tested in
controlled trials, the risks, particularly in terms of increased relapse, outweighed any
benefits (Dixon et al., 1995; Hirsch & Barnes, 1995).
The Schizophrenia Patient Outcomes Research Team (1998) concluded that

‘targeted, intermittent dosage maintenance strategies should not be used routinely in
lieu of continuous dosage regimens because of the increased risk of symptom
worsening or relapse. These strategies may be considered for patients who refuse
maintenance or for whom some other contraindication to maintenance therapy
exists, such as side-effect sensitivity’.
Clozapine
The antipsychotic clozapine was introduced in the1970s, only to be withdrawn soon
after because of the risk of potentially fatal agranulocytosis. However, after further
research revealed the drug’s efficacy in treatment-resistant schizophrenia (for
example, (Kane et al., 1988), clozapine was reintroduced in the 1980s with
requirements for appropriate haematological monitoring. Clozapine was considered
to have a novel mode of action. Its pharmacological profile includes a relatively low
affinity for D2 receptors and a much higher affinity for D4 dopamine receptors, and
for subtypes of serotonin receptors, although it is not clear exactly which aspects are
responsible for its superior antipsychotic effect in treatment-resistant schizophrenia.
Side effects
Clinical issues relating to side effects were summarised by (NICE, 2002a), as follows:
‘All antipsychotic agents are associated with side effects but the profile and
clinical significance of these varies among individuals and drugs. These may
include EPS (such as parkinsonism, acute dystonic reactions, akathisia and
tardive dyskinesia), autonomic effects (such as blurring of vision, increased
intra-ocular pressure, dry mouth and eyes, constipation and urinary
retention), increased prolactin levels, seizures, sedation and weight gain.
Cardiac safety is also an issue because several antipsychotics have been
shown to prolong ventricular repolarisation, which is associated with an
increased risk of ventricular arrhythmias. Routine monitoring is a pre-
requisite of clozapine use because of the risk of neutropenia and
agranulocytosis. Prescribers are therefore required to ensure that effective

ongoing monitoring is maintained as alternative brands of clozapine become
available.
Individuals with schizophrenia consider the most troublesome side effects to

be EPS, weight gain, sexual dysfunction and sedation. EPS are easily
recognised, but their occurrence cannot be predicted accurately and they are
related to poor prognosis. Akathisia is also often missed or misdiagnosed as
agitation. Of particular concern is tardive dyskinesia (orofacial and trunk
movements), which may not be evident immediately, is resistant to
treatment, may be persistent, and may worsen on treatment withdrawal.
Sexual dysfunction can be a problem, sometimes linked to drug-induced
hyperprolactinaemia; it is likely to be an underreported side effect of
antipsychotic treatment, as discussion of this issue is often difficult to
initiate.’
Blockade of D2 receptors by antipsychotic drugs is responsible for EPS, such as

parkinsonism, akathisia, dystonia and dyskinesia, but the therapeutic, antipsychotic
effect may occur at a lower level of D2 receptor occupancy than the level associated
with the emergence of EPS (Farde et al., 1992). SGA drugs were introduced with
claims for a lower risk of EPS. The individual SGAs differ in their propensity to
cause EPS: for some SGAs (for example, clozapine and quetiapine), acute EPS
liability does not differ from placebo across their full dose, while for some others the
risk is dose dependent. These differences may reflect individual drug profiles in
relation to properties such as selective dopamine D2-like receptor antagonism,
potent 5-HT2A antagonism and rapid dissociation from the D2 receptor, and for
aripiprazole, partial agonism at D2 and 5HT1A receptors. Interpretation of the RCT
evidence for the superiority of SGAs regarding acute EPS should take into account
the dosage and choice of FGA comparator, most commonly haloperidol, which is
considered a high potency D2 antagonist with a relatively high liability for EPS.
Raised serum prolactin is also an important adverse effect of antipsychotic

medication (Haddad & Wieck, 2004). It can lead to problems, such as menstrual
abnormalities, galactorrhea and sexual dysfunction, and in the longer term to
reduced bone mineral density (Haddad & Wieck, 2004; Meaney et al., 2004). While
the propensity for antipsychotic drugs to affect prolactin varies between agents, the
extent to which an individual service user will be affected may be difficult to
determine before treatment.
Antipsychotic drugs also have strong affinity for a range of other receptors, including
histaminergic, serotonergic, cholinergic and alpha-adrenergic types, which may
produce a number of other effects, such as sedation, weight gain and postural
hypotension. As the various antipsychotic drugs possess different relative affinities
for each receptor type, each drug will have its own specific profile of side effects. For
example, antipsychotic drugs vary in their liability for metabolic side effects, such as
weight gain, lipid abnormalities and disturbance of glucose regulation. These are side
effects that have been increasingly recognised as problems that may impact on long-

term physical health. Specifically, they increase the risk of the metabolic syndrome, a
recognised cluster of features (hypertension, central obesity, glucose
intolerance/insulin resistance and dyslipidaemia) (American Diabetes Association et
al., 2004; Mackin et al., 2007a), which is a predictor of type-2 diabetes and coronary
heart disease. Even without antipsychotic treatment, people with schizophrenia may
have an increased risk of such problems, which is partly related to lifestyle factors
such as smoking, poor diet, lack of exercise, and also, possibly, the illness itself.
(Brown et al., 1999; Holt et al., 2005; Osborn et al., 2007a; Osborn et al., 2007b; Taylor
et al., 2005; Van Nimwegen et al., 2008). While there is some uncertainty about the
precise relationship between schizophrenia, metabolic problems and antipsychotic
medication, there is agreement that routine physical health screening of people
prescribed antipsychotic drugs in the long term is required (Barnes et al., 2007;
Newcomer, 2007; Suvisaari et al., 2007) (further information about physical health
screening can be found in Chapter 7).
10.2 INITIAL TREATMENT WITH ANTIPSYCHOTIC

MEDICATION
10.2.1 Introduction
Evidence published before the 2002 guideline suggests that drug-naïve patients may
respond to doses of antipsychotic medication at the lower end of the recommended
range (Cookson et al., 2002; McEvoy et al., 1991; Oosthuizen et al., 2001; Tauscher &
Kapur, 2001). This may have particular implications in the treatment of people
experiencing their first episode of schizophrenia. Lehman and Steinwachs (1998)
have suggested that the maximum dose for drug-naïve patients should be 500 mg
chlorpromazine equivalents per day. This contrasts with a recommended optimal
oral antipsychotic dose of 300 to 1000 mg chlorpromazine equivalents per day for the
routine treatment of an acute episode in non-drug-naïve patients.

databases searched and the eligibility criteria can be found in Table 90. For the 2009
guideline, a new systematic search was conducted for relevant RCTs published since
the 2002 guideline (further information about the search strategy can be found in
Appendix 20).
10.2.3 Studies considered for review 34

Nine RCTs (N = 1,801) met the inclusion criteria for the 2009 guideline. Of these, two
trials (Emsley1995; Jones1998) were included in the 2002 guideline, but analysed
with the acute treatment trials (that is, non-initial treatment). All included studies
34Here and elsewhere in this chapter, each study considered for review is referred to by a study ID, with studies
included in the previous guideline in lower case and new studies in upper case (primary author and date or
study number for unpublished trials). References for included studies denoted by study IDs can be found in
Appendix 15b

are now published in peer-reviewed journals between 1999 and 2008. Further
22b.
10.2.4 Antipsychotic drug treatment in people with first-episode or

early schizophrenia
Of the nine RCTs included in the meta-analysis, two were multiple-arm trials and,
therefore, there were a total of 12 evaluations: three of olanzapine versus
haloperidol, one of olanzapine versus quetiapine, three of olanzapine versus
risperidone, four of risperidone versus haloperidol, and one of risperidone versus
quetiapine (see Table 91 for a summary of the study characteristics). Forest plots
and/or data tables for each outcome can be found in Appendix 23c.
Table 90: Clinical review protocol for the review of initial treatment with
antipsychotic medication
Primary clinical For people with first-episode or early schizophrenia, what are the benefits
question and downsides of continuous oral antipsychotic drug treatment when
compared with another oral antipsychotic drug at the initiation of treatment
(when administered within the recommended dose range [BNF54])?
Electronic data CENTRAL, CINAHL, EMBASE, MEDLINE, PsycINFO
bases
Study design Double-blind RCT (≥10 participants per arm and ≥4weeks’ duration)
Patient Adults (18+) with first-episode or early schizophrenia (including recent
population onset/people who have never been treated with antipsychotic medication)a
Excluded Very late on set schizophrenia (onset after age 60).

populations Other psychotic disorders, such as bipolar disorder, mania or depressive
psychosis.
difficulties, or substance misuse.
Interventions FGAs: SGAsb:

Benperidol Amisulpride
Chlorpromazine hydrochloride Aripiprazole
Flupentixol Olanzapine
Fluphenazine hydrochloride Paliperidone
Haloperidol Quetiapine
Levomepromazine Risperidone
Pericyazine Sertindole
Perphenazine Zotepine
Pimozide
Prochlorperazine
Promazine hydrochloride
Sulpiride
Trifluoperazine
Zuclopenthixolacetate
Zuclopenthixol dihydrochloride
Comparator Any relevant antipsychotic drug

Critical outcomes Mortality (suicide) Global state
(CGI)
Mental state (total symptoms,
depression) Social functioning
Leaving the study early for any
reason
Adverse events
Note. Studies (or outcomes from studies) were categorised as short term (12 weeks or fewer), medium term (12–
51weeks) and long term (52 weeks or more); studies that used drug doses outside the recommended dose range
were flagged during data analysis.
aStudies that included participants under the age of 18 were not excluded from the review unless all participants
were less than 18 years old.

bClozapine and sertindole were excluded from this analysis because they are not usually used to treat people
with first-episode or early schizophrenia.

Table 91: Summary of study characteristics for RCTs of antipsychotic drugs in people with first-episode or early
schizophrenia
Olanzapine Olanzapine Olanzapine Risperidone Risperidone versus

Versus Versus quetiapine Versus risperidone Versus haloperidol quetiapine
haloperidol
k (total N) 3 (331) 1 (267) 3 (446) 5 (1102) 1 (267)
Study ID DEHAAN2003 MCEVOY2007A Jones1998 Emsley1995 MCEVOY2007A
Jones1998 MCEVOY2007A Jones1998
LIEBERMAN2003A VANNIMWEGEN2008 LEE2007
MOLLER2008
SCHOOLER2005
Diagnostic DSM-IV DSM-IV DSM-IV DSM-III, DSM-IV DSM-IV
criteria
Baseline PANSS total:~81 PANSS total: PANSS total: PANSS total: PANSS total:
severity (SD15) Mean ~74 (SD ~16) mean~74 (SD 16) Range 77.3 to 94.2 Mean ~74 (SD 16)
(LIEBERMAN (MCEVOY2007A)
2003A)
Selected DEHAAN2003: Participants had to Jones1998: first 5 years of Emsley1995: first-episode Participants had to be in
inclusion 1–2psychotic be in first episode of illness Jones1998: first 5 years of first episode of
criteria episodes; aged their psychotic illness, MCEVOY2007A: illness; aged 18–65 years Their psychotic illness, and
17–28 years and had to be participants had to be in LEE2007: drug-naïve had to be continuously ill
Jones1998: first continuously ill for ≥1 first episode of their MOLLER2008: first for ≥1 month and no more
5 years of illness; month and nomore psychotic illness, and had episode; aged 18–60 than 5 months
aged 18–65 years than 5 months to be continuously ill for years
LIEBERMAN ≥ 1 month and no more
2003A: experienced than 5 months

Olanzapine Olanzapine Olanzapine Risperidone Risperidone versus

Versus Versus quetiapine Versus risperidone Versus haloperidol quetiapine
haloperidol
Psychotic symptoms VANNIMWEGEN2008: SCHOOLER2005:

for ≥1 month but not Recent onset; aged schizophrenia, <1year,
more than 60 18–30 years during which there were no
months; aged 16–40 more than two psychiatric
years hospitalisations for
psychosis and
≤ 12 weeks cumulative
exposure to antipsychotics;
Aged 16–45years
Age of DEHAAN2003: 16–44 years, mean Jones 1998: mean~29 years Emsley1995: 15–50 years, 16–44 years, mean 24.5 (SD5.8)
participants 17–26years 24.5 (SD5.8) MCEVOY2007A: 16–44 years, median~23years Jones1998: years
Jones 1998: mean mean 24.5 (SD 5.8) mean~29years
~29 years VANNIMWEGEN2008: mean LEE2007: mean 32.6 (SD1)
LIEBERMAN2003A: 25 years years
mean 23.9 (SD4.6) MOLLER2008: mean
30.1 (9.8) years
SCHOOLER2005: mean ~24
years
Setting Inpatient and Inpatient and outpatient Inpatient and outpatient Inpatient and outpatient Inpatient and outpatient
outpatient
Durationof Short term: 6 weeks Long term: 52 weeks Short term: 6 weeks Short term: 6–8weeks Long term: 52 weeks
treatment Medium term: Long term: 52–54 weeks Medium term: 24–30 weeks
12 weeks Long term: 54–104 weeks
Long term:
54–104 weeks
Medication dose Olanzapine: Olanzapine: Olanzapine: 2.5–20 (range) Risperidone: 2–10 (range) Risperidone: 0.5–4 (range)
(mg/day) 5–20 (range) 2.5–20 (range) Risperidone: 0.5–10 (range) Haloperidol: 1–20 (range) Quetiapine: 100–800 (range)
Haloperidol: 2.5–20 Quetiapine: 100–800
(range) (range)

In nine RCTs with a total of 1,801 participants with first-episode or early
schizophrenia (including people with a recent onset of schizophrenia and people
who have never been treated with antipsychotic medication), the evidence suggested
there were no clinically significant differences in efficacy between the antipsychotic
drugs examined. Most of the trials were not designed to examine differences in
adverse effects of treatment, but metabolic and neurological side effects reported
were consistent with those identified in the SPC for each drug.
10.3 ORAL ANTIPSYCHOTICS IN THE TREATMENT OF

THE ACUTE EPISODE
10.3.1 Introduction
Early clinical studies established that antipsychotic medications are effective in the
treatment of acute schizophrenic episodes (Davis & Garver, 1978 ), although they
proved to be more effective at alleviating positive symptoms than negative
symptoms, such as alogia or affective blunting. However, no consistent difference
between the FGAs was demonstrated in terms of antipsychotic efficacy or effects on
individual symptoms, syndromes or schizophrenia subgroups. Accordingly, the
choice of drug for an individual was largely dependent on differences in side-effect
profiles (Davis & Garver, 1978 ; Hollister, 1974). The limitations of these FGAs
included heterogeneity of response in acute episodes, with a proportion of
individuals showing little improvement (Kane, 1987) and a range of undesirable
acute and long-term side effects. The search for better-tolerated and more effective
drugs eventually generated a series of second-generation drugs, characterised by a
lower liability for EPS (Barnes & McPhillips, 1999; Cookson et al., 2002; Geddes et al.,
2000).

databases searched and the eligibility criteria can be found in Table 92. A new
systematic search for relevant RCTs, published since the 2002 guideline, was
conducted for the 2009 guideline (further information about the search strategy can
be found in Appendix 20).

Table 92: Clinical review protocol for the review of oral antipsychotics in the
treatment of the acute episode
Primary clinical For people with an acute exacerbation or recurrence of schizophrenia,

question what are the benefits and downsides of continuous oral antipsychotic
drug treatment when compared with another oral antipsychotic drug
(when administered within the recommended dose range [BNF 54])?
Electronic databases CENTRAL, CINAHL, EMBASE, MEDLINE, PsycINFO

Study design Double-blind RCT (≥10 participants per arm and ≥4 weeks’ duration)
Patient population Adults (18+) with an acute exacerbation or recurrence of schizophrenia
Excluded populations Very late onset schizophrenia (onset after age 60).
Other psychotic disorders, such as bipolar disorder, mania or
depressive psychosis.
People with coexisting learning difficulties, significant physical or
sensory difficulties, or substance misuse.
People with schizophrenia who have met established criteria for
treatment-resistant schizophrenia.
Interventions FGAs: SGAs 35:

Pericyazine Sertindole
Perphenazine Zotepine
Pimozide
Prochlorperazine
Sulpiride
Trifluoperazine
Zuclopenthixol acetate
Critical outcomes Mortality (suicide) Global state (CGI)
Mental state (total symptoms, depression) Social functioning
Adverse events
Note. Studies (or outcomes from studies) were categorised as short term (12 weeks or fewer),
medium term (12–51 weeks) and long term (52 weeks or more); studies that used drug doses
outside the recommended dose range were flagged during data analysis
.
35Clozapine was excluded from this analysis because it is not usually used to treat people with schizophrenia
unless criteria for treatment-resistant schizophrenia are met (see Section 10.5)

In the 2002 guideline, 180 RCTs were included 36. The search for the 2009 guideline
identified ten papers providing follow-up or published data for existing trials and 19
new trials. Two trials (Klieser1996; Malyarov1999) were multi-arm and contributed
to more than one comparison. Because of the large volume of evidence, the GDG
excluded open-label studies, head-to-head comparisons of two FGAs and
comparisons with placebo from the 2009 guideline review, leaving 72 RCTs (N =
16,556) that met inclusion criteria. Further information about both included and
excluded studies can be found in Appendix 22b.
10.3.4 Treatment with antipsychotic drugs in people with an acute

exacerbation or recurrence of schizophrenia
Because most included studies involved olanzapine or risperidone, comparisons
involving these drugs are reported first followed by comparisons involving other
drugs. Twenty-six RCTs compared olanzapine with another antipsychotic (see Table
93 for a summary of the study characteristics) and 30 compared risperidone with
another antipsychotic (see Table 94). Six RCTs were included in the analysis
comparing amisulpride with an FGA, two in the analysis compared aripiprazole
with an FGA and one compared aripiprazole with ziprasidone (see
Table 95); seven compared quetiapine with an FGA and two compared sertindole
with an FGA (see Table 96), and seven compared zotepine with an FGA (see Table
97). Forest plots and/or data tables for each outcome can be found in Appendix 23c.

In 72 RCTs involving 16,556 participants with an acute exacerbation or recurrence of
schizophrenia, there was little evidence of clinically significant differences in efficacy
between the oral antipsychotic drugs examined. Metabolic and neurological side
effects were consistent with those reported in the SPC for each drug.
36Of these, 146 trials came from the following existing sources: NICE TA43 (NICE, 2002) and the
Cochrane reviews of benperidol (Leucht & Hartung, 2002), loxapine (Fenton et al., 2002), pimozide
(Sultana & McMonagle, 2002), sulpiride (Soares et al., 2002) and thioridazine (Sultana et al., 2002).
New systematic reviews were conducted for chlorpromazine, flupentixol, fluphenazine, oxypertine,
pericyazine, perphenazine, prochlorperazine, promazine, trifluoperazine, and zuclopenthixol
dihydrochloride. Data from poor quality trials, placebo comparisons and drugs not available in the
UK were excluded

Table 93: Summary of study characteristics for olanzapine versus another antipsychotic drug (acute treatment)
Olanzapine versus Olanzapine versus Olanzapine versus Olanzapine versus

haloperidol another FGA amisulpride paliperidone
k (total N) 9 (3,071) 4 (249) 2 (429) 3 (1,090)
Study ID Beasley1996a HGBL1997 MARTIN2002 DAVIDSON2007
Beasley1997 Loza1999 WAGNER2005 KANE2007A
HGCJ1999 (HK) Jakovljevic1999 MARDER2007
HGCU1998 (Taiwan) Naukkarinen 1999/
Malyarov1999 HGBJ (Finland)
Reams1998
Tollefson1997
KONGSAKON2006
ROSENHECK2003
Diagnostic criteria DSM-III-R, DSM-IV, DSM-IV DSM-IV DSM-IV
Setting Inpatient and Inpatient and outpatient Inpatient and outpatient Inpatient and outpatient
outpatient
Duration of treatment Short term: 6 weeks Short term: 4–6 weeks Short term: 8 weeks Short term: 6 weeks
Medium term: Medium term: 26 weeks Medium term: 24 weeks
14–26 weeks
Long term: 52 weeks
Medication dose (mg/day) Olanzapine: 5–20 Olanzapine: 5–20 (range) Olanzapine: 5–20 (range) Olanzapine: 10 (range)
(range) Haloperidol: 5– Chlorpromazine Amisulpride: 200–800 Paliperidone: 6 or 9 kk
20 (range) hydrochloride: 200–800 (range)
(range)
Flupentixol: 5–20 (range)
Fluphenazine: 6–21 (range)
Perphenazine: 8–32 (range)
kkFor the purpose of the review, data from the 6 mg group (MARDER2007) and the 9 mg group (DAVIDSON2007) were used in the meta-analysis

Table 93: Summary of study characteristics for olanzapine versus another antipsychotic drug (acute treatment) (Continued)
Olanzapine versus quetiapine Olanzapine versus risperidone Olanzapine versus ziprasidone
k (total N) 1 (52) 5 (928) 2 (817)

Study ID RIEDEL2007B Conley2001 StudyR-0548 (SIMPSON2004)
Gureje1998 BREIER2005
Malyarov1999
Tran1997
STUDY-S036
Diagnostic criteria DSM-IV DSM-IV or ICD-10 DSM-IV
Setting Inpatient Inpatient and outpatient Inpatient and outpatient
Duration of treatment Short term: 8 weeks Short term: 6–8 weeks Short term: 6 weeks
Medium term: 26–30 weeks Medium term: 28 weeks
Medication dose Olanzapine: 15.82 (mean); Olanzapine: 5–20 (range) Olanzapine: 11.3–15.27 (range of
(mg/day) 10–20 (range) Risperidone: 2–12 (range) means)
Quetiapine: 586.86 (mean); Ziprasidone: 115.96–129.9 (range of
400–800 (range) means)

Table 94: Summary of study characteristics for risperidone versus another antipsychotic drug (acute treatment)
Risperidone versus Risperidone versus Risperidone versus Risperidone versus

haloperidol another FGA amisulpride aripiprazole
k (total N) 14 (2,437) 2 (205) 3 (585) 2 (487)
Study ID Blin1996 Hoyberg1993 Fleurot1997 CHAN2007B
Ceskova1993 Huttunen1995 Lecrubier2000 POTKIN2003A
Cetin1999 HWANG2003
Chouinard1993
Claus1991
Janicak1999
Liu2000
Malyarov1999
Marder1994
Mesotten1991
Min1993
Muller-Siecheneder1998
Peuskens1995
ZHANG2001
Diagnostic DSM-III-R, DSM-IV, ICD-9, DSM-III-R DSM-IV DSM-IV
criteria ICD-10
Setting Inpatient Not reported Inpatient Inpatient
Duration of Short term: 4–8 weeks Short term: 8 weeks Short term: 6–8 weeks Short term: 4 weeks
treatment Medium term: 12–26 weeks Medium term: 26 weeks
Medication dose Risperidone: 5.5–12 (range of Risperidone: 8–8.5 (range of Risperidone: 4–10 (range) Risperidone: 6 (fixed)
(mg/day) means); 1–20 (range) means); 15–20 (max) Amisulpride: 400–1000 Aripiprazole: 15, 20, 30 (fixed)
Haloperidol: 9.2–20 (range of Perphenazine: 28 (mean); 48 (range)
means); 2–20 (range) (max) Zuclopenthixol: 38
(mean); 100 (max)

Table 94: Summary of study characteristics for risperidone versus another antipsychotic drug (acute treatment) (Continued)
Risperidone versus Risperidone versus Risperidone versus Risperidone versus zotepine

quetiapine sertindole ziprasidone
k (total N) 1 (673) 1 (187) 1 (296) 1 (59)
Study ID ZHONG2006 AZORIN2006 Study128-302 Klieser1996
(ADDINGTON2004)
Diagnostic DSM-IV DSM-IV DSM-III-R ICD-9
criteria
Setting Inpatient and outpatient Inpatient and outpatient Not reported Not reported
Duration of Short term: 8 weeks Medium term: 12 weeks Short term: 8 weeks Short term: 4 weeks
treatment
Medication dose Risperidone: 6.0 (mean); Risperidone: 6.6 (mean); 4–10 Risperidone: 7.4 (mean); Risperidone: 4 or 8 (fixed)
(mg/day) 2–8 (range) (range) 3–10 (range) Zotepine: 225 (fixed)
Quetiapine: 525 (mean); Sertindole: 16.2 (mean); Ziprasidone: 114
200–800 (range) 12–24 (range) (mean);

Table 95: Summary of study characteristics for amisulpride or aripiprazole versus another antipsychotic drug (acute treatment)
Amisulpride versus Amisulpride versus Aripiprazole versus Aripiprazole versus

haloperidol another FGA haloperidol ziprasidone
k (total N) 5 (921) 1 (132) 2 (1,708) 1 (256)
Study ID Carriere2000 Hillert1994 KANE2002 ZIMBROFF2007
Delcker1990 KASPER2003
Moller1997
Puech1998
Ziegler1989
Diagnostic DSM-III-R, DSM-IV, ICD-9 DSM-III-R DSM-IV DSM-IV
criteria
Setting Inpatient and outpatient Inpatient Inpatient and outpatient Inpatient and outpatient
Duration of Short term: 4–6 weeks Short term: 6 weeks Short term: 4 weeks Short term: 4 weeks
treatment Medium term: 16 weeks Long term: 52 weeks
Medication dose Amisulpride: 400–2,400 Amisulpride: 956 (mean); Aripiprazole: 15 or 30 (fixed) Aripiprazole: 20.9 (mean
(mg/day) (range) 1000 (maximum) Haloperidol: 10 (fixed) modal)
Haloperidol: 10–40 (range) Flupentixol: 22.6 (mean); Ziprasidone: 149 (mean
25 (maximum) modal)

Table 96: Summary of study characteristics for quetiapine or sertindole versus an
FGA (acute treatment)
Quetiapine versus Quetiapine versus Sertindole versus

haloperidol another FGA haloperidol
k (total N) 4 (818) 1 (201) 1 (617)
Study ID Arvanitis1997 Link1994 Hale2000
Fleischhacker1996
Purdon2000
ATMACA2002
Diagnostic DSM-III-R, DSM-IV, DSM-III-R DSM-III-R
criteria ICD-10
Setting Inpatient and outpatient Not reported Inpatient
Duration of Short term: 6 weeks Short term: 6 weeks Short term: 8 weeks
treatment Medium term: 26 weeks
Medication Quetapine: 50–800 Quetapine: 407 (mean) Sertindole: 8, 16 or
dose (range) Haloperidol: 1– Chlorpromazine 20, 24 (fixed)
(mg/day) 16 (range) hydrochloride: Haloperidol: 10 (fixed)
384 (mean)
Table 97: Summary of study characteristics for zotepine versus an FGA (acute
treatment)
Zotepine versus haloperidol Zotepine versus another

FGA
k (total N) 5 (386) 2 (146)
Study ID Barnas1987 Cooper1999a
Fleischhacker1989 Dieterle1999
Klieser1996
Petit1996
KnollCTR (StudyZT4002)
Diagnostic DSM-III, DSM-III-R, ICD-9 DSM-III-R, ICD-9
criteria
Setting Inpatient Mostly inpatient
Duration of Short term: 4–8 weeks Short term: 4–8 weeks
treatment Medium term: 26 weeks
Medication dose Zotepine: 94–309 (range Zotepine: 241 (mean); 300 (max)
(mg/day) of means); 150–300 (range) Chlorpromazine
Haloperidol: 4–15 (range hydrochloride: 600 (max)
of means); 10–20 (range) Perphenazine: 348 (mean)

10.4 PROMOTING RECOVERY IN PEOPLE WITH
SCHIZOPHRENIA THAT ARE IN REMISSION –
PHARMACOLOGICAL RELAPSE PREVENTION
10.4.1 Introduction
Following their introduction into clinical practice in the early 1950s, chlorpromazine
and related drugs rapidly became widely used for both acute treatment of people
experiencing symptoms of psychosis and for prevention of relapse. By the 1980s,
haloperidol (synthesised in 1959) became the most widely used drug for these
purposes in the US (Davis et al., 1993; Gilbert et al., 1995; Healy, 2002; Hirsch &
Barnes, 1995). A meta-analysis (Davis et al., 1993) of 35 double-blind studies
compared maintenance treatment using FGAs with placebo in over 3,500 service
users. Relapse was reported in 55% of those who were randomised to receive
placebo, but in only 21% of those receiving active drugs. Gilbert et al. (1995)
reviewed 66 antipsychotic withdrawal studies, published between 1958 and 1993,
and involving over 4,000 service users. The mean cumulative rate of relapse in the
medication withdrawal groups was 53% (follow-up period 6 to 10 months)
compared with 16% (follow-up of 8 months) in the antipsychotic maintenance
groups. Over a period of several years, continuing treatment with conventional
antipsychotics appears to reduce the risk of relapse by about two-thirds (Kissling,
1991).
When the effects of stopping antipsychotic drugs after an acute psychotic episode or
after long-term maintenance treatment were examined, the subsequent rate of
relapse seemed to be similar in both situations. Individuals who are well stabilised
on maintenance medication show high rates of relapse when their antipsychotic
therapy is discontinued (Kane, 1990) or switched to placebo (Hogarty et al., 1976). A
recent Cochrane review (Almerie et al., 2007 ) including ten trials of chlorpromazine
cessation in stable participants (total N = 1,042) showed that those stopping
chlorpromazine had a relative risk of relapse in the short term (up to 8 weeks) of 6.76
(95% CI, 3.37 to 13.54) and in the medium term (9 weeks to 6 months) of 4.04 (95%
CI, 2.81 to 5.8). Relative risk of relapse after 6 months was 1.70 (95% CI, 1.44 to 2.01).
Another meta-analysis of data from several large collaborative studies (Davis et al.,
1993) suggested that the number of people who survive without relapse after
discontinuing drug treatment declines exponentially by around 10% a month.
Whether maintenance drug treatment is required for all people with schizophrenia is
uncertain. Around 20% of individuals will only experience a single episode (Möller
& van Zerssen, 1995). A recent pragmatic observational study analysing over 4,000
participants who achieved remission in the Schizophrenia Outpatient Health
Outcomes study, showed that 25% relapsed over a 3-year follow-up period with a
constant rate of relapse over this time (Haro et al., 2007). It therefore appears that a
proportion of people will experience a relapse despite continued antipsychotic drug
treatment. It is unclear whether such people benefit from an increase in antipsychotic
dosage during episodes of psychotic exacerbation (Steingard et al., 1994).

Given that there are no consistent reliable predictors of prognosis or drug response,
the 2009 guideline, as well as other consensus statements and guidelines, generally
recommend that pharmacological relapse prevention is considered for every patient
diagnosed with schizophrenia (for example (Lehman et al., 1998) and (Dixon et al.,
1995). Possible exceptions are people with very brief psychotic episodes without
negative psychosocial consequences, and the uncommon patient for whom all
available antipsychotics pose a significant health risk (Fleischhacker & Hummer,
1997).
It is clear from the placebo-controlled RCTs and discontinuation studies cited above
that the efficacy of antipsychotics in relapse prevention is established. However, it is
also clear from recent pragmatic trials that switching of medication over time is
common in clinical practice (Jones et al., 2006; Lieberman et al., 2005). In the Clinical
Antipsychotic Trials of Intervention Effectiveness (CATIE) study (Lieberman et al.,
2005), 74% of participants discontinued their randomised treatment over 18 months
(further information about this trial can be found in Section 10.8 on the effectiveness
of antipsychotic medication). This may well reflect the need in clinical practice to
search collaboratively for the drug that offers the best balance of efficacy and
tolerability for the individual patient. The role of depot preparations in contributing
to concordance and continuation on medication is discussed in Section 10.6.
All the antipsychotics identified for review have established supremacy over placebo
in the prevention of relapse, although the evidence that any individual antipsychotic
drug, or group of antipsychotics (FGAs and SGAs), has greater efficacy or better
tolerability than another is still very uncertain. One of the main aims of antipsychotic
drug development in recent decades has been to produce compounds with
equivalent antipsychotic efficacy, but without troubling EPS. The doses of
haloperidol that came to be used in routine clinical practice by the 1980s and early
1990s were higher than those required for its antipsychotic effect, and EPS were
common. The trials conducted in the 1990s comparing SGAs and haloperidol often
tested the latter at relatively high doses, arguably above the optimum for at least a
proportion of the subjects treated, and highlighted the propensity of haloperidol to
cause such side effects in comparison with SGAs. The widespread introduction of
SGAs to clinical practice from the mid1990s onwards thus appeared to offer a
genuine therapeutic advance. However, more recent effectiveness (pragmatic) trials
have suggested that the claimed advantages of these drugs may have been
overstated, especially if their propensity to cause metabolic abnormalities and other
side effects is taken into account, and if they are compared with FGAs (other than
higher dose haloperidol) (Geddes et al., 2000; Jones et al., 2006; Lieberman et al.,
2005; NICE, 2002a). SGAs are not a homogeneous class and may not deserve a group
title. They differ widely in their pharmacology and side effect profile. There are
unanswered questions regarding their relative efficacy and tolerability and their use
over the long-term compared with FGAs. Their risks of long-term metabolic
disturbance are not yet fully quantified and neither is the risk of movement
disorders, such as tardive dyskinesia compared with FGAs, so any small advantage

that may be offered by reduced EPS may be offset by these other adverse
consequences not shown by the earlier drugs.
While evaluating each drug against each other would appear superficially the best
way of approaching the question posed for this review, in reality the number of
possible comparisons and the limited number of studies available would render this
a meaningless task. Therefore, the GDG considered that comparing the individual
SGAs against all FGA comparators, primarily in terms of relapse, provided the most
meaningful analysis of the available data.
Definitions
The definitions of relapse used in this review were those adopted by the individual
studies. This definition varied between studies (see Sections 10.4.4 and 10.4.5), and
therefore, caution should be exercised in the interpretation of the results.

databases searched and the eligibility criteria used for this section of the guideline
can be found in Table 98. A new systematic search for relevant RCTs, published
since the 2002 guideline, was conducted for the 2009 guideline (further information
about the search strategy can be found in Appendix 20 and information about the
search for health economic evidence can be found in Section 10.9.1).

In the 2002 guideline, nine RCTs comparing an SGA with an FGA were included
(based on a then unpublished review by Leucht and colleagues). Leucht and
colleagues published their review in 2003; it included one additional trial and six
trials comparing an SGA with placebo that were not included in the 2002 guideline.
For the 2009 guideline, the review was limited to double-blind RCTs of
antipsychotics used for relapse prevention; therefore, four studies (Daniel1998;
Essock1996; Rosenheck1999; Tamminga1994) included in the 2002 guideline were
excluded from the 2009 guideline review. In addition, one trial of an SGA versus
another SGA, included in the 2002 review of acute treatment, met the criteria for
inclusion in this review (Tran1997). The search for the 2009 guideline identified four
additional RCTs (one comparing an SGA with an FGA, one comparing an SGA with
an SGA, and one comparing an SGA with placebo). For the purposes of the health
economic model (see Section 10.9.2), trials of ziprasidone versus placebo were
included because this drug has been compared with a licensed SGA.
In total, 17 RCTs (N = 3,535) met the inclusion criteria for the 2009 guideline review.
Of these, one was unpublished (STUDY-S029) and the remainder were published in
peer- reviewed journals between 1994 and 2007. Further information about both
included and excluded studies can be found in Appendix 22b.
Table 98: Clinical review protocol for the review of relapse prevention
Primary clinical For people with schizophrenia that is in remission, what are the
question benefits and down sides of continuous oral antipsychotic drug
treatment when compared with another antipsychotic drug (when
administered within the recommended dose range [BNF54])?

Study design Double-blind RCT (≥10 participants per arm and ≥ 6 months’
duration)
Patient population Adults (age 18+) with schizophrenia that is in remission (for the
purposes of the guideline, remission includes people who have
responded fully or partially to treatment)
Excluded populations Very late onset schizophrenia (onset after age 60).
Other psychotic disorders, such as bipolar disorder, mania or
depressive psychosis. People with coexisting learning difficulties,
significant physical or sensory difficulties, or substance misuse.

Interventions FGAs: SGAs:
Pericyazine Zotepine
Perphenazine
Pimozide
Prochlorperazine
Sulpiride
Trifluoperazine
Comparator Any relevant antipsychotic drug or placebo
Critical outcomes Global state (relapse).
Overall treatment failure (relapse or leaving the study early for
any reason).
Leaving the study early because of adverse events.
Note. Studies (or outcomes from studies) were categorised as short term (12 weeks or
fewer), medium term (12–51 weeks) and long term (52 weeks or more); studies that used
drug doses outside the recommended dose range were flagged during data analysis
10.4.4 Second-generation antipsychotics versus placebo in people with

schizophrenia that is in remission (relapse prevention)
Eight RCTs were included in the meta-analysis comparing an SGA (amisulpride,
aripiprazole, olanzapine, paliperidone, ziprasidone, zotepine) with placebo (see
Table 99). Forest plots and/or data tables for each outcome can be found in
Appendix 23c.

Table 99: Summary of study characteristics for of an SGA versus placebo (relapse prevention)
Amisulpride versus Aripiprazole versus placebo Olanzapine versus placebo

placebo
k (total N) 1 (141) 1 (310) 3 (446)
StudyID LOO1997 PIGOTT2003 BEASLEY2000
DELLVA1997(study1)
DELLVA1997(study2)
Selected inclusion Residual or disorganised Chronic schizophrenia with BEASLEY2000 a
criteria schizophrenia; predominant diagnosis made at least 2 years DELLVA1997(studies 1and 2) b
negative symptoms prior to entry and continued
antipsychotic treatment during
this period
Diagnostic criteria DSM-III-R DSM-IV DSM-III-R
Definition of relapse Withdrawal because of Impending decompensation BEASLEY2000: Hospitalisation for positive symptoms or
inefficacy of treatment and based on one or more of the ≥4 increase on BPRS positive score or increase of single
PANSS > 50 following: a CGI-I ≥ 5; a PANSS ≥ BPRS item to 4 and increase from baseline ≥2
5 on subscore items of hostility or
uncooperativeness on 2 DELLVA1997: Hospitalisation for psychopathology
successive days; or a ≥ 20%
increase in PANSS total score
Duration o 26 weeks 26 weeks 42– 46 weeks
ftreatment
Setting Outpatient Inpatient and outpatient Outpatient

Medication dose Amisulpride: 100 (fixed) Aripiprazole: 15 (fixed) BEASLEY2000, olanzapine: 10–20 (range)
(mg/day) DELLVA1997, olanzapine: ~12 (semi-fixed)

Paliperidone versus Ziprasidone versus placebo Zotepine versus placebo
placebo
k (total N) 1 (207) 1 (277) 1 (119)
Study ID KRAMER2007 ARATO2002 COOPER2000
Selected inclusion Achieved stabilisation after Lack of acute relapse, lack of Rating of at least mildly ill according to CGI; relapse in
criteria 8-week hospitalisation for an treatment resistance, and living the 18 months before inclusion
acute episode, then further under medical supervision for at
6-week stabilisation least 2 months
Diagnostic criteria DSM-IV DSM-III-R DSM-III-R

Definition of relapse Recurrent episode of Hospitalisation for Hospitalisation for psychopathology
schizophrenia psychopathology
Duration of treatment 46 weeks 52 weeks 26 weeks
Setting Inpatient initially, then Inpatient Inpatient/outpatient

outpatient
Medication dose Palperidone: 10.8 (mean); Ziprasidone: 40, 80 or 160 (fixed) Zotepine: 150 or 300 (fixed)
(mg/day) 3–15 (range)
Note. a Minimally symptomatic; negative symptoms; at least 6 weeks of stability; continued stability while taking olanzapine during an 8-week
period.
b Responder from 6-week acute treatment phase (responders defined as ≥40% reduction in BPRS score or BPRS score ≤18).

10.4.5 Second-generation antipsychotics versus another antipsychotic
drug in people with schizophrenia that is in remission (relapse
prevention)
Nine RCTs were included in the meta-analysis comparing an SGA (amisulpride,
olanzapine, risperidone) with an FGA (haloperidol) (see Table 100), and two were
included in the analysis comparing an SGA (olanzapine) with another SGA
(risperidone, ziprasidone) (see Table 101). Forest plots and/or data tables for each
outcome can be found in Appendix 23c.

In 17 RCTs including 3,535 participants with schizophrenia, the evidence suggested
that, when compared with placebo, all of the antipsychotics examined reduced the
risk of relapse or overall treatment failure. Although some SGAs show a modest
benefit over haloperidol, there is insufficient evidence to choose between
antipsychotics in terms of relapse prevention.
10.5 PROMOTING RECOVERY IN PEOPLE WITH

SCHIZOPHRENIA WHOSE ILLNESS HAS NOT
RESPONDED ADEQUATELY TO TREATMENT
10.5.1 Introduction
The phrase ‘treatment-resistant’ is commonly used to describe people with
schizophrenia whose illness has not responded adequately to treatment. The essence
of treatment resistance in schizophrenia is the presence of poor psychosocial and
community functioning that persists despite trials of medication that have been
adequate in terms of dose, duration and adherence. While treatment resistance is
sometimes conceptualised in terms of enduring positive psychotic symptoms, other
features of schizophrenia can contribute to poor psychosocial and community
functioning, including negative symptoms, affective symptoms, medication side
effects, cognitive deficits and disturbed behaviour. Treatment resistance in
schizophrenia is relatively common, in that between a fifth and a third of service
users show a disappointing response to adequate trials of antipsychotic medication
(Brenner et al., 1990; Conley & Buchanan, 1997; Lieberman et al., 1992). In a small
proportion of people experiencing their first episode of schizophrenia, the illness
will be resistant to antipsychotic medication, showing only a limited response (for
example, precluding early discharge from hospital) (Lambert et al., 2008; Lieberman
et al., 1989; Lieberman et al., 1992; MacMillan et al., 1986; May, 1968 ), but more
commonly the illness becomes progressively more unresponsive to medication over
time (Lieberman et al., 1993; Wiersma et al., 1998).

Table 100: Summary of study characteristics for RCTs of an SGA versus another antipsychotic drug (relapse prevention)
Amisulpride versus Olanzapine versus haloperidol Risperidone versus haloperidol

haloperidol
K (total N) 1 (60) 4 (1082) 2 (428)
StudyID Speller1997 Tran1998a Csernansky2000
Tran1998b MARDER2003 a
Tran1998c
STUDY-S029
Selected inclusion Chronic, long-term Tran1998(a,b,c): Responder from a Csernansky2000: Stability according
criteria hospitalised inpatient; 6-week acute treatment (at least 40% to clinical judgment; receipt of the
moderate to severe negative reduction of BPRS score or BPRS score same medication for 30days; same
symptoms ≤18) residence for 30 days
STUDY-S029: Received a stable dose of MARDER2003: Atleast two acute
the same conventional antipsychotic episodes in last 2 years or 2 years of
drug ≥8weeks before visit 1; had a continuing symptoms; receipt of
PANSS score ≥49 at visit 2; considered treatment as an outpatient for at least
as possible patient in the patients with 1 month
schizophrenia study (that is, patient
global outcome improvement or
benefit, such as optimisation of long-
term therapy) who should benefit from
a switch of current therapy based on
investigator’s judgment as a result of
efficacy (PANSS score ≥ 49) or
tolerability concerns.
Diagnostic criteria DSM-III-R, DSM-IV DSM-IV

Amisulpride versus Olanzapine versus haloperidol Risperidone versus haloperidol

haloperidol
Definition of relapse Increase of three or more Tran1998(a,b,c): Hospitalisation for Csernansky2000: (1) Hospitalisation;
BPRS positive symptom items psychopathology (2) increase of level of care and 20%
that did not respond to a dose increase in PANSS score; (3) self-injury,
increase STUDY-S029: Psychiatric suicidal or homicidal ideation,
hospitalisation or 25% increase in the Violent behaviour; (4)CGI rating >6
PANSS total score in relation to
baseline or major deterioration in MARDER2003: Increase >3 in the
clinical condition defined by a CGI-I BPRS scores for the thought disorder
score of 6 or 7, or suicide attempt that and hostile-suspiciousness clusters,
required medical treatment and/or or an increase > 2 in the score for
jeopardised vital prognosis either of these clusters and as core >3
on at least one item of these clusters
Duration of 52 weeks 22–84 weeks 52 weeks

treatment
Setting Inpatient Inpatient/outpatient Outpatient
Medication dose Amisulpride: 100–800; Tran1998 a and b Risperidone: ~5 (mean);
(mg/day) Haloperidol: 3–20 b Olanzapine: ~12 (semi-fixed) 2–16 (range)
Haloperidol: ~14 (semi-fixed) Haloperidol: <5–12 (range of means);
2–20 (range)
Tran1998c
Olanzapine: 14 (mean);
5–20 (range)
Haloperidol: 13 (mean); 5–20 (range)
Note. a Duration was 2 years, but 1-year data was used for the review to enhance comparability
b
A minimum effective dose strategy was followed.

Table 101: Summary of study characteristics for RCTs of an SGA versus
another SGA (relapse prevention)
Olanzapine versus risperidone Olanzapine versus ziprasidone

K (total N) 1 (339) 1 (126)
Study ID Tran1997 SIMPSON2005
Selected inclusion Minimum BPRS of 42 and excluded for Responders to 6-week acute treatment
criteria failure to show minimal clinical response trial of olanzapine or risperidone
with antipsychotics in three chemical (response defined as a CGI-I of ≤2 or a
classes dosed at ≥ 800 chlorpromazine ≥20% reduction in PANSS at acute-study
hydrochloride equivalents/day or end point, and outpatient status)
clozapin edosed at ≥400mg/day for at
least 6weeks
Diagnostic criteria DSM-IV DSM-IV
Definitionof relapse 20% or greater worsening in the PANSS ≥ 20% worsening of PANSS total score
total score along with a CGI-S score ≥ 3 and a CGI severity score ≥3
after 8 weeks of therapy
Durationof 28 weeks 28 weeks
treatments
Setting Inpatient or outpatient Outpatient
Medication dose Olanzapine: 17.2 (mean modal); Olanzapine: 12.6 (mean); 5–15 (range)
(mg/day) 10–20 (range) Ziprasidone: 135.2 (mean);
Risperidone: 7.2 (mean modal); 78–162 (range)
4–12 (range)
The definition of the term ‘treatment-resistant schizophrenia’ varies considerably in

the studies covered in this review. Kane et al. (1988) introduced rigorous
criteria involving aspects of the clinical history, cross-sectional measures and
prospective assessments. One trend has been a move towards broader definitions of
treatment resistance that allow a larger number of individuals to be viewed as
clinically eligible for treatment with clozapine. For example, Bondolfi et al. (1998)
included in their trial people with chronic schizophrenia who ‘had previously failed
to respond to or were intolerant of at least two different classes of antipsychotic
drugs given in appropriate doses for at least 4 weeks each’. Others have adopted an
even wider clinical notion of ‘incomplete recovery’ (Pantelis & Lambert, 2003), which
acknowledges the presence of lasting disability in functional and psychosocial
aspects despite psychological/psychosocial and pharmacological interventions,
while also recognising the potential for improvement.
10.5.2 Treatment-resistant schizophrenia and antipsychotic medication

High-dosage antipsychotic medication is commonly used for treatment-resistant
schizophrenia, although there is little evidence to suggest any significant benefit
with such a strategy (Royal College of Psychiatrists, 2006). Clinicians may also try
switching to another antipsychotic, although similarly the research evidence on the
possible value of such a strategy is not consistent or promising (Kinon et al., 1993;
Lindenmayer et al., 2002; Shalev et al., 1993). An alternative strategy has been to try

to potentiate antipsychotics by combining them either with each other (see Section
10.5.3) or with other classes of drugs. Possible adjuncts to antipsychotic treatment
include mood stabilisers and anticonvulsants, such as lithium, carbamazepine,
sodium valproate, lamotrigine, antidepressants and benzodiazepines (Barnes et al.,
2003; Chong & Remington, 2000; Durson & Deakin, 2001). However, the use of such
adjunctive treatments to augment the action of antipsychotics is beyond the scope of
this guideline.
Kane and colleagues (1988; 2001) established the efficacy of clozapine over FGAs in
strictly-defined treatment-resistant schizophrenia, and subsequent meta- analyses
have confirmed the superiority of clozapine in terms of reducing symptoms and the
risk of relapse (Chakos et al., 2001; Wahlbeck et al., 1999). However, Chakos et al.
(2001) concluded from their meta-analysis that the evidence for clozapine when
compared with the SGAs tested was inconclusive. Even with optimum clozapine
treatment, the evidence suggests that only 30 to 60% of treatment-resistant
schizophrenia will show a satisfactory response (Iqbal et al., 2003). As clozapine is
associated with severe and potentially life-threatening side effects, particularly the
risk of agranulocytosis, the SPC states that drug should only be considered where
there has been a lack of satisfactory clinical improvement despite adequate trials, in
dosage and duration, of at least two different antipsychotic agents including an SGA.
Monitoring plasma clozapine concentration may be helpful in establishing the

optimum dose of clozapine in terms of risk–benefit ratio, and also in assessing
adherence (Gaertner et al., 2001; Llorca et al., 2002; Rostami-Hodjegan et al., 2004)
particularly for service users showing a poor therapeutic response or experiencing
significant side effects despite appropriate dosage. An adequate trial will involve
titrating the dosage to achieve a target plasma level, usually considered to be above
350mg/l, although response may be seen at lower levels (Dettling et al., 2000;
Rostami-Hodjegan et al., 2004). If the response to clozapine monotherapy is poor,
augmentation strategies may be considered (see Section 10.5.3 for a review of the
evidence).
A number of patient-related factors have been reported to increase the variability of

plasma clozapine concentrations, with gender, age and smoking behaviour being the
most important (Rostami-Hodjegan et al., 2004). Smoking is thought to increase the
metabolism of clozapine by inducing the cytochrome P450 1A2 (CYP1A2) and other
hepatic enzymes (Flanagan, 2006; Ozdemir et al., 2002). The metabolism of clozapine
is mainly dependent on CYP1A2. This has several clinical implications. First, there is
some evidence that smokers are prescribed higher doses by clinicians to compensate
for higher clozapine clearance (Tang et al., 2007). Secondly, plasma concentrations of
clozapine and its active metabolite, norclozapine, vary considerably at a given
dosage, and this variation may be greater in heavy smokers receiving lower doses of
clozapine, increasing the risk of subtherapeutic concentrations (Diaz et al., 2005).
Thirdly, prompt adjustment of clozapine dosage in patients who stop smoking
during treatment is important, to avoid the substantially elevated clozapine

concentrations and increased risk of toxicity that would otherwise be expected
(Flanagan, 2006; McCarthy, 1994; Zullino et al., 2002).
10.5.3 Combining antipsychotic drugs

In clinical practice, the prescription of combined antipsychotics is relatively
common. A multi-centre audit of the prescription of antipsychotic drugs for
inpatients in 47 mental health services in the UK, involving over 3,000 inpatients,
found that nearly half were receiving more than one antipsychotic drug (Harrington
et al., 2002). Similarly, prescription surveys in the UK by Taylor and colleagues
(2000; 2002) and the Prescribing Observatory for Mental Health (Paton et al., 2008)
have confirmed a relatively high prevalence of combined antipsychotics for people
with schizophrenia, including co-prescription of FGAs and SGAs.
The reasons for such prescriptions include as required (‘p.r.n.’) medication, a

gradual switch from one antipsychotic drug to another and adding an oral
antipsychotic to depot treatment to stabilise illness. A common rationale for
combining antipsychotics is to achieve a greater therapeutic response when there has
been an unsatisfactory response to a single antipsychotic. In this respect, there is
little supportive evidence for superior efficacy (Chan & Sweeting, 2007; Chong &
Remington, 2000), and Kreyenbuhl and colleagues (2007) reported that psychiatrists
perceive antipsychotic polypharmacy to be generally ineffective for persistent
positive psychotic symptoms. The concerns with combined antipsychotics include
prescribing higher than necessary total dosage and an increased risk of side effects. If
there is clinical benefit, one problem is the attribution of this to the combination
rather than one or other of the individual antipsychotics, and thus uncertainty about
the implications for optimal pharmacological treatment longer term.
For treatment-resistant schizophrenia that has proved to be unresponsive to

clozapine alone, adding a second antipsychotic would seem to be a relatively
common strategy. The prevalence of this augmentation strategy in people with
schizophrenia on clozapine ranges from 18 to 44% depending on the clinical setting
and country (Buckley et al., 2001; Potter et al., 1989; Taylor et al., 2000).
The mechanisms that might underlie any increase in therapeutic effect with
combined antipsychotics have not been systematically studied (Mccarthy &
Terkelsen, 1995). However, in relation to the strategy of adding an antipsychotic to
clozapine, it has been hypothesised that any pharmacodynamic synergy might be
related to an increased level of D2 dopamine receptor occupancy, above a threshold
level (Chong & Remington, 2000; Kontaxakis et al., 2005). However, such an increase
might also be expected to be associated with an increased risk of EPS. An alteration
of the interaction between serotonin (5-hydroxytryptamine) and D2 activity has also
been suggested as a relevant mechanism (Shiloh et al., 1997). Further,
pharmacokinetic interactions might play a part, although there is no consistent
evidence that adding an antipsychotic leads to increased clozapine plasma levels
(Honer et al., 2006; Josiassen et al., 2005; Yagcioglu et al., 2005).

RCTs and open studies have reported clozapine augmentation with a second
antipsychotic to be relatively well tolerated. The main treatment-emergent side
effects have been predictable from the pharmacology of the augmenting drug, with
EPS and prolactin elevation among the most common problems. However, with
risperidone as the augmenting antipsychotic there are isolated reports of problems
such as agranulocytosis, a trial ectopics and possible neuroleptic malignant
syndrome (Chong et al., 1996; Godleski & Sernyak, 1996; Kontaxakis et al., 2002);
with aripiprazole as the second antipsychotic, there are reports of nausea, vomiting,
insomnia, headache and agitation in the first 2 weeks (Ziegenbein et al., 2006) and
also modest weight loss (Karunakaran et al., 2006; Ziegenbein et al., 2006).

The clinical review protocol, including the primary clinical questions, information
about the databases searched and the eligibility criteria, can be found in Table 102. A
new systematic search for relevant RCTs, published since the 2002 guideline, was
Table 102: Clinical review protocol for the review of interventions for people
with schizophrenia whose illness has not responded adequately to treatment
Primary clinical questions For people with schizophrenia whose illness has not responded adequately
to treatment, what are the benefits and downsides of continuous oral
antipsychotic drug treatment when compared with another antipsychotic
drug (when administered within the recommended dose range [BNF54])?
For people with schizophrenia with persistent negative symptoms, what are
the benefits and downsides of continuous oral antipsychotic drug treatment
when compared with another antipsychotic drug (when administered
within the recommended dose range [BNF54])?
For people with schizophrenia whose illness has not responded adequately
to clozapine treatment, is augmentation of clozapine with another
antipsychotic associated with an enhanced therapeutic response?

Study design Double-blind RCT (≥10 participants per arm and ≥4 weeks’ duration)
Patient population Adults (18+) with schizophrenia whose illness has not responded
adequately to treatment (including those with persistent negative
symptoms
Excluded populations Very late onset schizophrenia (onset after age 60). Other psychotic
disorders, such as bipolar disorder, mania or depressive psychosis.

Interventions FGAs: Benperidol SGAs: Amisulpride
Flupentixol Clozapine
Fluphenazine hydrochloride Olanzapine
Haloperidol Paliperidone
Levomepromazine Quetiapine
Pericyazine Risperidone
Perphenazine Sertindole
Pimozide Zotepine
Prochlorperazine
Sulpiride
Trifluoperazine

Global state (relapse)
Mental state (total symptoms, negative symptoms, depression)
Social functioning
Cognitive functioning
Adverse events
Note. Studies (or outcomes from studies) were categorised as short term (12 weeks or fewer),
medium term (12–51 weeks) and long term (52 weeks or more); studies that used drug doses outside
the recommended dose range were flagged during data analysis.
a Studies that only included participants with persistent negative symptoms were analysed
separately.

In the 2002 guideline, 19 RCTs were included in the review of antipsychotic
medication for people with schizophrenia whose illness has not responded
adequately to treatment. The search for the 2009 guideline identified five papers
providing follow-up data or published versions of existing trials, and eight new
trials (one trial [LIBERMAN2002] provided no useable outcome data and was
excluded from the analysis). In addition, six trials (Altamura1999; Breier2000;
Conley1998a; Emsley1999; Heck2000; Kern1998) analysed in the 2002 guideline as
acute phase studies were now included in the 2009 review, and three (Essock1996a;
Gelenberg1979b; Wahlbeck2000) previously included in the 2002 guideline were
excluded in the 2009 guideline. In total, 26 trials (N = 3,932) met the inclusion criteria
for the 2009 guideline review. Further information about both included and
excluded studies can be found in Appendix 22b.
A new analysis, not conducted for the 2002 guideline, examined RCTs of
antipsychotic medication in people with persistent negative symptoms of

schizophrenia. Three trials (Boyer1990; Lecrubier1999; Murasaki1999) included in
the 2002 review of acute treatment are now included here, but excluded from the
review of acute treatment in the 2009 guideline. One trial (OLIE2006 1) excluded from
the 2002 guideline is now included. One trial (Speller1997) included in the relapse
prevention review also met the inclusion criteria for this review. The search for the
2009 guideline also identified five new RCTs that are included in this review, and
one trial (HERTLING2003) that reported no appropriate data and so was excluded
from the analysis. In total, ten RCTs (N =1,200) met the inclusion criteria for the 2009
guideline review. Further information about both included and excluded studies can
be found in Appendix 22b.
For the review of clozapine augmentation, an existing systematic review and meta-
analysis (Paton et al., 2007), published since the 2002 guideline, was used as the basis
for an updated meta-analysis in the 2009 guideline. This published review focused
on the augmentation of clozapine with another SGA and included four RCTs. The
search for the 2009 guideline identified two further RCTs. In total, six trials (N = 252)
met the inclusion criteria for the update. In addition, two small studies (Assion et al.,
2008; Mossaheb et al., 2006) with fewer than ten participants in either arm were
excluded, and one trial of clozapine plus amisulpride versus clozapine plus
quetiapine (Genc et al., 2007) was excluded. Further information about both included
and excluded studies can be found in Appendix 22b.
10.5.6 Clozapine versus another antipsychotic drug in people with

schizophrenia whose illness has not responded adequately to
treatment
Seven RCTs were included in the analysis comparing clozapine with an FGA in
people with schizophrenia whose illness has not responded adequately to treatment
(see Table 103), and ten RCTs were included in the analysis of clozapine versus
another SGA (see Table 104). Forest plots and/or data tables for each outcome can be
found in Appendix 23c.
1 In the previous guideline this trial this was labelled as ‘Study 128-305’.

Table 103: Summary of study characteristics for RCTs of clozapine versus an FGA in people with schizophrenia whose
illness has not responded adequately to treatment
Clozapine versus haloperidol Clozapine versus anon-haloperidol FGA

K (total N) 4 (607) 3 (459)
Study ID Buchanan1998 Claghorn1987
Klieser1989 Hong1997
Rosenheck1997 Kane1988
VOLAVKA2002
Diagnostic criteria DSM-III-R, DSM-IV DSM-II, DSM-III, DSM-IV
Selected inclusion Buchanan1998: Non-complete response to at least two trials of Claghorn1987: In tolerant to at least two prior antipsychotics
criteria therapeutic doses of antipsychotics for at least 6 weeks Hong1997: Treatment-refractory (severe psychotic symptoms
Klieser1989: Chronic treatment-resistant (no diagnostic criteria) according to BPRS item scores for> 6 months despite
Rosenheck1997: Treatment-resistant, high level use of inpatient treatment with antipsychotics from at least two different
services classes at dosages of at least 1000 mg chlorpromazine
VOLAVKA2002: Suboptimal response to previous treatment, hydrochloride equivalents)
defined by history of persistent positive symptoms after at least 6 Kane1988: ≥3 periods of antipsychotic treatment,
contiguous weeks of treatment with one or more typical 1000mg/day of chlorpromazine hydrochloride equivalents
antipsychotics at ≥600mg/d in chlorpromazine hydrochloride without significant symptomatic relief and BPRS total score of
equivalents, and a poor level of functioning over past 2 years at least 45
Setting Inpatient/outpatient Inpatient

Duration of Short term: 6–10 weeks Short term: 4–8weeks
treatment Medium term: 14 weeks Medium term: 12weeks
Long term: 52weeks
Medication dose Clozapine: 400–552mg/day (range of means); Clozapine: 417–543mg/d (range of means);
(mg/day) 100–900mg/day (range) 150–900mg/d (range)
Haloperidol: 20–28mg/day (range of means); Chlorpromazine hydrochloride: 798–1163mg/day (range of
5–30mg/day (range) means); 300–1800mg/day (range)
Note. aAll three trials used chlorpromazine as the comparator.

Table 104: Summary of study characteristics for RCTs of clozapine versus another SGA in people with schizophrenia whose
illness has not responded adequately to treatment
Clozapine versus olanzapine Clozapine versus risperidone Clozapine versus zotepine
K (total N) 5 (485) 5 (529) 1 (50)

Study ID Beuzen1998 Anand1998 Meyer-Lindberg
Bitter1999 (BITTER2004) Bondolfi1998 1996
MELTZER2008 Breier1999
Oliemeulen2000 Chowdhury1999
VOLAVKA2002 VOLAVKA2002
Diagnostic criteria DSM-IV DSM-III-R, DSM-IV, ICD-10 DSM-III-R
Selected inclusion Beuzen1998: Treatment resistant, >3 Anand1998: Treatment resistant: severe, Unresponsive to >3 weeks of two
criteria on at least two items of PANSS chronic disease and poor response to FGAs ineffective doses, BPRS>39
positive subscale previous antipsychotics (no period of
Bitter1999: Treatment-resistant or good functioning for at least 24 months
intolerant individuals must have not despite the use of two antipsychotics,
responded adequately to standard current episode without significant
acceptable antipsychotic medication, improvement for at least 6 months
either because of ineffectiveness or despite the use of an antipsychotic
because of intolerable side effects equivalent to haloperidol 20mg for at
caused by the medication least 6 weeks, total
MELTZER2008: Documented history BPRS at least 45, and CGI at least 4
of treatment-resistant schizophrenia Bondolfi1998: Treatment resistant:
based on Kane and colleagues’ (1988) failed to respond/intolerant to >2
criteria different classes of antipsychotics in
Oliemeulen2000: Therapy-resistant; appropriate doses for >4 weeks
schizophrenia or other psychotic Breier1999: Partial response to
disorders antipsychotics, defined as a history of

VOLAVKA2002: Suboptimal residual positive and/or negative
response to previous treatment, symptoms after at least a 6-week trial of
defined by history of persistent a therapeutic dose of a antipsychotic
positive symptoms after at least 6 and at least a minimum level of
contiguous weeks of treatment with symptoms
one or more typical antipsychotics at Chowdhury1999: Duration of illness
≥600 mg/day in chlorpromazine >6 months and received at least one full
hydrochloride equivalents, and a course of FGA without adequate
poor level of functioning over past 2 response, or cases intolerant to FGAs
years because of intractable neurological and
non-neurological side effects,
necessitating withdrawal of drug or
inadequate dosing VOLAVKA2002: see
left
Setting Inpatient/outpatient Inpatient (not stated in three trials) Not stated
Duration of treatment Short term: 8 weeks Short term: 6–8 weeks Short term: 6 weeks
Medium term: 14–26 weeks Medium term: 12–16 weeks
Medication dose Clozapine: 564 mg/day (mean); Clozapine: 291–597.5 mg/d (range of Clozapine: 150–450 mg/day (range)
(mg/day) 200–900 mg/day (range) Olanzapine: means); 150–900 mg/d (range) Zotepine: 150–450 mg/d (range)
33.6 mg/day (mean); Risperidone: 5.8–8.3 mg/day (range of
10–45 mg/day (range) means); 2–16 mg/day (range)

10.5.7 Second-generation antipsychotic drugs (other than clozapine)
versus first-generation antipsychotic drugs in people with
treatment
Ten RCTs were included in the analysis comparing clozapine with another
antipsychotic in people with schizophrenia whose illness has not responded
adequately to treatment (see Table 105). Forest plots and/or data tables for each

versus second-generation antipsychotic drugs in people with
treatment
Three RCTs were included in the analysis comparing an SGA (olanzapine and
risperidone) with another SGA in people with schizophrenia whose illness has not
responded adequately to treatment (see Table 106). Forest plots and/or data tables
for each outcome can be found in Appendix 23c.

versus another antipsychotic in people who have persistent
negative symptoms
Five RCTs were included in the analysis comparing an SGA (amisulpride,
olanzapine, quetiapine, risperidone) with another SGA in people who have
persistent negative symptoms (see Table 107). Five RCTs were included in the
analysis comparing an SGA (amisulpride, olanzapine, quetiapine, risperidone) with
another SGA in people who have persistent negative symptoms (see Table 108).
Forest plots and/or data tables for each outcome can be found in Appendix 23c.
10.5.10 Combining antipsychotics (augmentation of clozapine

with another second-generation antipsychotic drug)
One trial was included in the analysis comparing clozapine plus aripiprazole with
clozapine plus placebo, four trials compared clozapine plus risperidone with
clozapine plus placebo, and one trial compared clozapine plus sulpiride with
clozapine plus placebo (see Table 109). Forest plots and/or data tables for each

Table 105: Summary of study characteristics for RCTs of SGAs versus FGAs in people with schizophrenia whose illness has
not responded adequately to treatment
Aripiprazole versus a non- Olanzapine versus haloperidol Olanzapine versus a non-haloperidol FGA
haloperidol FGA
K (total N) 1 (300) 3 (617) 1 (84)
Study ID KANE2007B Altamura1999 Conley1998a
(ALTAMURA2002)
Breier2000
BUCHANAN2005
Diagnostic criteria DSM-IV DSM-IV DSM-III-R
Selected inclusion criteria Treatment resistant (defined as failure Altamura1999: Partial or non- responders Treatment resistant: Non-responders during
to experience satisfactory symptom to treatment according to preset criteria haloperidol phase.
relief despite at least two periods of Breier2000: Sub-population from
treatment, each lasting ≥6 weeks with Tollefson1997with treatment- resistant
adequate doses of antipsychotics) schizophrenia, defined as failure to
respond to at least one neuroleptic over a
period of at least 8 weeks during the
previous 2 years
BUCHANAN2005: Partial response to
fluphenazine during 4-week open-label
phase
Setting Inpatient/outpatient Inpatient/outpatient Inpatient

Duration of treatment Short term: 6weeks Short term: 6weeks Short term: 8weeks
Medium term: 14–16weeks
Medication dose Aripiprazole: 15–30mg/day (range) Olanzapine: 11.1–12.4mg/day (range of Olanzapine: 25mg/day (fixed)
(mg/day) Perphenazine: 8–64mg/day (range) means); 5–30mg/day (range) Chlorpromazine hydrochloride: 1200mg/day
Haloperidol: 10–12.3mg/day (range of (fixed)
means); 5 30mg/day (range)

Table 105: Summary of study characteristics for RCTs of SGAs versus FGAs in people with schizophrenia whose illness has not
responded adequately to treatment (Continued)
Quetiapine versus Quetiapine versus a Risperidone versus haloperidol Risperidone versus a non-haloperidol FGA
haloperidol non-haloperidol FGA
K (total N) 1 (288) 1 (25) 3 (161) 1 (26)
Study ID Emsley1999 CONLEY2005 Heck2000 CONLEY2005
Kern1998
SEE1999
Diagnostic DSM-IV DSM-IV DSM-III-R, DSM-IV DSM-IV
criteria
Selected inclusion Persistent positive Treatment resistanta Heck2000: Disturbing EPS during their Treatment resistant
criteria symptoms while previous neuroleptic treatment
previously taking Kern1998: Treatment resistant
antipsychotics according to the Kane criteria
SEE1999: A history of partial
responsiveness to FGAs and residual
symptoms
Setting Not reported Inpatient Not reported Inpatient
Duration of Short term: 8weeks Medium term: 12weeks Short term: 5–8 weeks Medium term: 12 weeks
treatment
Medication dose Quetiapine: Quetiapine: 400mg/day Risperidone: 7mg/day (mean) Risperidone: 4mg/day (fixed)
(mg/day) 600mg/day (fixed) (fixed) (Kern1998); 16mg/day (max) Fluphenazine hydrochloride: 12.5mg/day
Haloperidol: 20mg/day Fluphenazine (Heck2000) (fixed)
(fixed) hydrochloride: Haloperidol: 19mg/day (mean)
12.5mg/day (fixed) (Kern1998); 24mg/day (max)
(Heck2000)
Note. a Defined by: (1) Persistent positive symptoms (≥4 points on 2 of 4 BPRS psychosis items); (2) Persistent global illness severity (BPRS total ≥45 and CGI
≥4); (3) At least two prior failed treatment trials with two different antipsychotics at doses of ≥600 mg/day chlorpromazine hydrochloride equivalent each of
at least 6 weeks’ duration; (4) No stable period of good social/occupational functioning in past 5 years.

Table 106: Summary of study characteristics for RCTs of SGAs versus SGAs in people with schizophrenia whose illness has
not responded adequately to treatment
Olanzapine versus Olanzapine versus ziprasidone Risperidone versus quetiapine

risperidone
k (total N) 1 (80) 1 (394) 1 (25)

Study ID VOLAVKA2002 KINON2006A CONLEY2005
Diagnostic criteria DSM-IV DSM-IV DSM-IV
Selected inclusion Suboptimal response to previous Prominent depressive symptoms Treatment resistant c
criteria treatment a
Setting Inpatient Outpatient Inpatient
Duration of treatment Medium term: 14 weeks Medium term: 24 weeks Medium term: 12 weeks
Medication dose Olanzapine: 10–40mg/day Olanzapine: 10, 15 or 20mg/day Risperidone: 4mg/day (fixed) Quetiapine:
(mg/day) (range) (fixed) 400mg/day (fixed)
Risperidone: 4–16mg/day Ziprasidone: 80, 120 or 160mg/day
(range) (fixed)
Note. a Defined by history of persistent positive symptoms after at least 6 contiguous weeks of treatment with one or more typical antipsychotics at
≥600mg/day chlorpromazine hydrochloride equivalent, and a poor level of functioning over past 2 years.
b Defined by a MADRS score≥16 (mild depression) and a score ≥4 (pervasive feelings of sadness or gloominess) on item 2 (reported sadness) of the
MADRS.
c Defined by: (1) Persistent positive symptoms (≥4 points on 2 of 4 BPRS psychosis items); (2) Persistent global illness severity (BPRS total ≥45 and
CG I≥ 4); (3) At least two prior failed treatment trials with two different antipsychotics at doses of ≥600 mg/day chlorpromazine hydrochloride
equivalent each of at least 6 weeks’duration; (4) No stable period of good social/occupational functioning in past 5 years.

Table 107: Summary of study characteristics for RCTs of SGAs versus a FGA in people who have persistent negative
symptoms
Amisulpride versus Amisulpride versus Olanzapine versus Quetiapine versus Risperidone versus
haloperidol A non-haloperidol haloperidol Haloperidol a non-haloperidol
FGA FGA
K (total N) 1 (60) 1 (62) 1 (35) 1 (197) 1 (153)
Study ID Speller1997 Boyer1990 LINDENMAYER2007 Murasaki1999 RUHRMANN2007
Diagnostic criteria Not reported DSM-III DSM-IV DSM-IV or ICD-10 ICD-10
Selected inclusion Chronic, long-term All met Andreasen Fulfilled criteria for Predominantly Negative symptoms
criteria hospitalised criteria for negative the Schedule for the negative symptoms (≥3 on
inpatients with symptoms and Deficit Syndrome PANSS negative
moderate to severe absence of marked (SDS) which included subscale)
negative symptoms positive symptoms. negative symptoms
that are stable rather
than unstable-state
manifestations
Setting Not reported Not reported Inpatient/outpatient Inpatient/outpatient Inpatient/outpatient

Duration of treatment Long term: 52 weeks Short term: 6 weeks Medium term: 12 Short term: 8weeksMedium term: 25
weeks weeks
Medication dose Amisulpride: Amisulpride: Olanzapine: 15– Quetiapine: Risperidone: 2–
(mg/day) 100–800mg/day 225mg/day (mean); 20mg/day 226mg/day (mean); 6mg/day
Haloperidol: 50–300mg/day (range) 600mg/day (max) (range)
3–20mg/day (range) Haloperidol: 15– Haloperidol: Flupentixol: 4–
Fluphenazine 20mg/day 6.7mg/day (mean); 12mg/day
hydrochloride: (range) 18mg/day (max) (range)
10mg/day (mean);
2–12mg/day (range)

Table 108: Summary of study characteristics for RCTs of SGAs versus another SGA in people who have persistent negative
symptoms
Amisulpride versus Olanzapine versus Olanzapine versus Risperidone versus quetiapine

ziprasidone amisulpride quetiapine
K (total N) 1 (123) 1 (140) 2 (386) 1 (44)
Study ID OLIE2006 Lecrubier1999 KINON2006B RIEDEL2005
(LECRUBIER2006) SIROTA2006
Diagnostic criteria DSM-III-R DSM-IV DSM-IV DSM-IV or ICD-10
Selected inclusion criteria Negative symptoms Primarily negative Prominent negative Predominantly primary negative
(baseline scores on the symptoms according to symptoms according to symptoms according to PANSS.
PANSS negative subscale PANSS and SANS PANSS and GAF/SANS.
had to exceed the PANSS
positive subscale by ≥6)
Setting Outpatient Inpatient/outpatient Inpatient/outpatient Inpatient/outpatient

Duration of treatment Medium term: 12weeks Medium term: 26weeks Medium term: 12–26 Medium term: 12 weeks
weeks
Medication dose (mg/day) Amisulpride: Olanzapine: 5 or Olanzapine: 5–20mg/day Risperidone:
144.7mg/day (mean); 20mg/day (fixed) (range) 4.9mg/day (mean);
100–200mg/day (range) Amisulpride: 150mg/day Quetiapine: 200– 2–6mg/day (range)
Ziprasidone: 118mg/day (fixed) 800mg/day (range) Quetiapine:
(mean); 589.7mg/day (mean);
80–160mg/day (range) 50–600mg/day (range)

Table 109: Summary of study characteristics for trials of clozapine augmentation
Clozapine+aripiprazole versus Clozapine+risperidone versus Clozapine+sulpiride versus

clozapine+placebo clozapine+placebo clozapine+placebo
K (total N) 1 (62) 4 (162) 1 (28)

Study ID CHANG2008 FREUDENREICH2007 SHILOH1997
HONER2006
JOSIASSEN2005
YAGCIOGLU2005
Diagnostic criteria DSM-IV DSM-IV DSM-IV
Inclusion criteria (1) Failure to respond to at least FREUDENREICH2007: (1) Failure (1) DSM diagnosis of
two previous antipsychotic drugs; to respond to at least two previous schizophrenia; (2) Clozapine
(2) Clozapine treatment for more antipsychotics; (2) currently prescribed after failure to respond
than 1 year with at least 8 weeks at treated with clozapine to three typical antipsychotics at
a stable daily dose of 400 mg or monotherapy for at least 6 adequate doses for at least 6weeks
more, unless compromised by months, at a stable dose for at least each; (3) 25 or more on BPRS; (4)
adverse effects; 8 weeks and with clozapine BPRS scores table for 5 weeks; (5)
(3) No change in clozapine daily plasma levels of at least Inability to function as an
dose or other concomitant 200ng/mL, unless the clozapine outpatient
medication for more than 3 dose necessary to achieve that
months, indicating a plateau of level was not tolerated
clinical response to clozapine;
(4) Either a baseline BPRS total HONER2006: (1) DSM diagnosis
score of at least 35 or more than of schizophrenia; (2) 80 or more on
two SANS global rating item PANSS and 4 or more on CGI; 3)
scores of at least 3 40 or less on Social and
Occupational Functioning
Assessment Scale; 4) Failure to
respond (≥20% reduction in BPRS)
after one placebo augmentation
for 1 week

JOSIASSEN2005: 1) DSM
diagnosis of schizophrenia; 2)
Continued significant psychotic
symptoms; 3) Failure to respond to
at least two previous antipsychotic
drugs; 4) 45 or more on BPRS or 4
or more (moderately ill) on at least
two BPRS positive symptoms
subscale items (hallucinatory
behaviour, conceptual
disorganisation, unusual thought
content, suspiciousness)
YAGCIOGLU2005: 1) DSM
diagnosis of schizophrenia; 2)
Failure to respond to at least two
previous antipsychotic drugs;
3) 72 or more on PANSS or 4 or
more on
CGI (moderate level of
psychopathology);
4) Prescribed clozapine because of
failure to respond to other
antipsychotic treatments
Setting Inpatient/outpatient Inpatient/outpatient Inpatient
Baseline severity BPRS total 47.6 (clozapine + Range of means: PANSS total BPRS total 41.9
aripiprazole)/48.5 (clozapine + 72.4–102.5 (clozapine + (clozapine + sulpiride)/43.5
placebo) risperidone)/73.5–97.8 (clozapine (clozapine + placebo)
+ placebo)
Duration of treatment 8 weeks FREUDENREICH2007: 6weeks 10 weeks

HONER2006: 8 weeks
JOSIASSEN2005: 12 weeks
YAGCIOGLU2005: 6 weeks

In 18 RCTs including 2,554 participants whose illness had not responded adequately
to treatment, clozapine had the most consistent evidence for efficacy over the FGAs
included in the trials. Further evidence is required to establish equivalence between
clozapine and any other SGA, and to establish whether there are differences between
any of the other antipsychotic drugs. Side effects were consistent with those reported
in the SPC for each drug.
In 10 RCTs including 1,200 participants with persistent negative symptoms, there

was no evidence of clinically significant differences in efficacy between any of the
antipsychotic drugs examined. Careful clinical assessment to determine whether
such persistent features are primary or secondary is warranted, and may identify
relevant treatment targets, such as drug-induced parkinsonism, depressive features
or certain positive symptoms.
In six RCTs including 252 participants with schizophrenia whose illness had not
responded adequately to clozapine treatment, there was some evidence that
clozapine augmentation with a second antipsychotic might improve both total and
negative symptoms if administered for an adequate duration.
10.6 TREATMENT WITH DEPOT/ LONG-ACTING

INJECTABLE ANTIPSYCHOTIC MEDICATION
10.6.1 Introduction
The introduction of long-acting injectable formulations (‘depot’) of antipsychotic
medication in the 1960s was heralded as a major advance in the treatment of
established schizophrenia outside hospital. At the time it was hoped that depot
preparations would lead to improved outcomes from antipsychotic
pharmacotherapy. Consistent drug delivery and avoidance of the bioavailability
problems that occur with oral preparations (such as gut wall and hepatic first-pass
metabolism) were felt to be important factors. Other benefits include eliminating the
risk of deliberate or inadvertent overdose. In the subsequent decades, the main
practical clinical advantage to emerge has been the avoidance of covert non-
adherence (both intentional and unintentional) 1 to antipsychotic drug treatment,
where there is close nursing supervision and documentation of clinic attendance
(Barnes & Curson, 1994; Patel & David, 2005). Service users who are receiving depot
treatment and who decline their injection or fail to receive it (through forgetfulness
or any other reason) can be immediately identified; allowing appropriate
1Further information about medicines concordance and adherence to treatment can be found in the NICE guideline on this
topic (seehttp://www.nice.org.uk).

intervention, bearing in mind that poor adherence to the medication can be both a
cause and consequence of worsening illness. In practice, the use of depot drugs does
not guarantee good treatment adherence, with a significant number who are
prescribed maintenance treatment with depot preparations after discharge from
hospital failing to become established on the injections (Crammer & Eccleston, 1989;
Young et al., 1999; Young et al., 1986). But for those who continue with long-acting
injections, there may be some adherence advantage over oral antipsychotics,
indicated by a longer time to medication discontinuation (Zhu et al., 2008). There is
also some evidence to suggest a better global outcome with depot as compared with
oral antipsychotics (Adams et al., 2001) with a reduced risk of rehospitalisation
(Schooler, 2003; Tiihonen et al., 2006). In 2002, a long-acting formulation of an SGA,
risperidone, became available, offering the same advantages of convenience and the
avoidance of covert non-adherence (Hosalli & Davis, 2003.).
Information on the use of long-acting antipsychotic injections has been limited

(Adams et al., 2001), but relevant surveys and audits of antipsychotic prescription in
the UK suggest that between a quarter and a third of psychiatric patients prescribed
an antipsychotic may be receiving a long-acting injection, depending on the clinical
setting (Barnes et al., 2009; Foster et al., 1996; Paton et al., 2003).
10.6.2 Use of long-acting antipsychotic injections

Long-acting injectable antipsychotic formulations generally consist of an ester of the
drug in an oily solution. Another way of formulating such a preparation is to use
microspheres of the drug suspended in aqueous solution. These drugs are
administered by deep intramuscular injection and are then slowly released from the
injection site, giving relatively stable plasma drug levels over long periods, allowing
the injections to be given every few weeks. However, this also represents a potential
disadvantage because there is a lack of flexibility of administration, with adjustment
to the optimal dosage being a protracted and uncertain process. The controlled
studies of low-dose maintenance treatment with depot preparations suggest that any
increased risk of relapse consequent upon a dose reduction may take months or
years to manifest. Another disadvantage is that, for some people, receiving the depot
injection is an ignominious and passive experience. Further, there have been reports
of pain, oedema, pruritus and sometimes a palpable mass at the injection site. In
some people, these concerns may lead service users to take active steps to avoid
these injections and even disengage with services altogether rather than receive
medication via this route. Nevertheless, a substantial proportion of people receiving
regular, long-acting antipsychotic injections prefer them to oral therapy, largely
because they consider them to be more convenient (Patel & David, 2005; Walburn et
al., 2001).

The review protocol, including the primary clinical questions, information about the
databases searched and the eligibility criteria, can be found in Table 110. A new
systematic search for relevant RCTs, published since the 2002 guideline, was

Table 110: Clinical review protocol for the review of depot/long-acting injectable
antipsychotics
Primary clinical For people with schizophrenia that is in remission, is any depot or long-
questions acting antipsychotic medication associated with improved relapse prevention
overtime?
For people with schizophrenia whose illness has not responded adequately
to treatment and who have had long-term antipsychotic drug treatment,is
there any evidence that patients have a preference for either depot/long-
acting or oral preparations?

Study design Double-blind RCT (≥10 participants per arm and ≥4 weeks’duration)
Patient population Adults (18+) with schizophrenia

Excluded populations Very late onset schizophrenia (onset after age 60). Other psychotic disorders,
such as bipolar disorder, mania or depressive psychosis.
Interventions FGAs:
Flupentixol decanoate
Fluphenazine decanoate
Haloperidol (as decanoate)
Pipotiazine palmitate
Zuclopenthixol decanoate
SGAs:
Risperidone (long-acting injection)
Comparator Any relevant antipsychotic drug or placebo
Globalstate (CGI, relapse)
Mental state (total symptoms, negative symptoms, depression)
Social functioning
Adverse events
Note. Studies (or outcomes from studies) were categorised as short term (12 weeks or fewer), medium
term (12–51 weeks) and long term (52 weeks or more).

In the 2002 guideline, the review of depot antipsychotic medication was based on a
meta-review of five Cochrane reviews (David & Adams, 2001), which included 13

RCTs of flupentixol decanoate, 48 of fluphenazine decanoate, 11 of haloperidol
decanoate, ten of pipothiazine palmitate and three of zuclopenthixol decanoate.
Since publication of the 2002 guideline, the review of fluphenazine decanoate (David
& Adams, 2001) was updated and now includes 70 trials. The review of pipothiazine
palmitate (Dinesh et al., 2004) was also updated and now includes 18 trials. In
addition, one SGA (long-acting injectable risperidone) has been licensed for use as a
depot. A Cochrane review of this medication for people with schizophrenia was
published in 2003 (Hosalli & Davis, 2003.). The search for the 2009 guideline
identified no additional trials that met the eligibility criteria. Because of the volume
of evidence for FGA depots, the GDG checked the updated Cochrane reviews were
consistent with the 2002 guideline and then focused on the evidence for long-acting
risperidone, which had not previously been reviewed. In total, two trials (N = 1,042)
met inclusion criteria (one trial of long-acting risperidone versus placebo, and one
trial of long- acting risperidone versus oral risperidone). Both trials were published
in peer- reviewed journals between 2003 and 2005. Further information about the
included studies can be found in Appendix 22b.
10.6.5 Long-acting risperidone injection versus placebo or oral

risperidone
One RCT was included in the analysis comparing long-acting risperidone injection
with placebo injection, and one RCT was included in the analysis comparing long-
acting risperidone with oral risperidone plus placebo injection (see Table 111). Forest
plots and/or data tables for each outcome can be found in Appendix 23c.

The search for the 2009 guideline did not identify any new evidence for the efficacy
and safety of depot FGAs beyond that included in the updated Cochrane reviews
(utilised in the 2002 guideline). These reviews did not indicate robust new evidence
that would warrant changing the existing recommendations for depot antipsychotic
medication.
Since publication of the 2002 guideline, the first depot SGA (risperidone) was
licensed for use in the UK. However, there is currently only limited evidence from
two double-blind RCTs regarding the efficacy and safety of long-acting injectable
risperidone compared with placebo or oral antipsychotic medication (risperidone).
The placebo controlled trial suggests that 25–75 mg of long-acting risperidone may
improve the chance of response and produce a clinically significant reduction in the
symptoms of schizophrenia, but larger doses carry an increased risk of neurological
side effects. There is no evidence to suggest that long-acting risperidone has either
greater efficacy or greater risk of adverse effects when compared with oral
risperidone. However, as suggested by the trial authors, the trial was only designed
to investigate the short-term switching of participants from oral medication to long-
acting risperidone; further studies are needed to understand the effect of continuous
delivery of this medication.

Table 111: Summary of study characteristics for RCTs of long-acting risperidone versus placebo or oral risperidone
Intramuscular injection of long-acting Intramuscular injection of long-acting risperidone

risperidone versus placebo injection versus oral risperidone+ placebo injection
K (total N) 1 (400) 1 (642)

Study ID KANE2003 CHUE2005
Diagnostic criteria Schizophrenia (DSM-IV) Schizophrenia (DSM-IV)
Baseline severity 25mg long-acting risperidone: PANSS total: Long-acting risperidone: PANSS total: mean 68.4 (SD
Mean 81.7 (SD 12.5), n = 99 1.0), n = 319
50mg long-acting risperidone: PANSS total: Oral risperidone: PANSS total:
Mean 82.3 (SD 13.9), n = 103 Mean 69.3 (SD 0.9), n = 321
75mg long-acting risperidone: PANSS total:
Mean 80.1 (SD 14.0), n = 100 All participants were required to be symptomatically
Placebo: stable during the last 4 weeks of the run-in period
PANSS total: mean 82.0 (SD 14.4), n = 98
Run-in 1-week oral risperidone run-in period 8 weeks open-label period during which participants were
stabilised on oral risperidone
Setting Inpatient/outpatient Inpatient/outpatient
Duration of treatment 12 weeks 12 weeks
Medication dose Fixed dose of 25, 50 or 75 mg every 2 weeks Long-acting risperidone: 88 participants received
(mg/day) 25mg every 2 weeks, 126 received 50mg and 105 received
75mg
Oral risperidone: 86 participants received 2mg/day,

126 received 4mg/day and 109 received 6mg/day

10.7 SIDE EFFECTS OF ANTIPSYCHOTIC MEDICATION
10.7.1 Introduction
Given that for some antipsychotics there was a paucity of side-effect data, the GDG
decided to pool data, where appropriate, from the studies included in the other
meta- analyses reported in this chapter and from any other relevant clinical trial. The
review focused on metabolic and neurological side effects as these were considered a
priority by the GDG and were also highlighted as areas of concern by service users.

All RCTs included in the efficacy reviews (except studies of depot/long-acting
antipsychotics) were included in the overall side effects meta-analysis. In addition,
four trials (ATMACA2003; LIEBERMAN2003B; MCQUADE2004; MELTZER2003)
did not meet the inclusion criteria for any of the efficacy reviews, but reported
relevant side effect data and so were included here.
10.7.3 Second-generation antipsychotic drugs versus another

antipsychotic drug (overall analysis of side effects)
As shown in Table 112, 14 separate RCTs were included in the analysis of
amisulpride against haloperidol (k = 6), a non-haloperidol FGA (k = 2), or an SGA (k
= 6). Seven separate trials were included in the analysis of aripiprazole against
haloperidol (k = 2), a non-haloperidol FGA (k = 1), or an SGA (k = 4). Sixteen
separate trials were included in the analysis of clozapine against haloperidol (k = 4),
a non-haloperidol FGA (k = 4), or an SGA (k = 9). Forty-one separate trials were
included in the analysis of olanzapine against haloperidol (k = 18), a non-haloperidol
FGA (k = 5), or an SGA (k = 19). Three trials were included in the analysis of
paliperidone against an SGA (k = 3). Thirteen separate trials were included in the
analysis of quetiapine against haloperidol (k = 5), a non-haloperidol FGA (k = 2), or
an SGA (k = 7). Forty separate trials were included in the analysis of risperidone
against haloperidol (k = 20), a non-haloperidol FGA (k = 4), or an SGA (k = 18).
Three separate trials were included in the analysis of sertindole against haloperidol
(k = 2), or an SGA (k = 1). Seven separate trials were included in the analysis of
zotepine against haloperidol (k = 5), a non-haloperidol FGA (k = 1), or an SGA (k =
1). Forest plots and/or data tables for each outcome can be found in Appendix 23c.

Table 112: Summary of studies included in the overall analysis of side effects
Treatment Comparator
Versus haloperidol (FGA) Versus non-haloperidol FGA Versus SGA
Amisulpride Carriere2000 [16weeks] Boyer1990 (fluphenazine) [6 weeks] Fleurot1997 (risperidone) [8 weeks]
Delcker1990 [6 weeks] Hillert1994 (flupentixol) [6 weeks] HWANG2003 (risperidone) [6 weeks]
Moller1997 [6 weeks] Lecrubier1999 (olanzapine) [26 weeks]
Puech1998[4 weeks] Lecrubier2000 (risperidone) [26 weeks]
Speller1997 [52 weeks] MARTIN2002 (olanzapine) [24 weeks]
Ziegler1989 [4 weeks] WAGNER2005 (olanzapine) [8 weeks]
k=6 k=2 k=6

Aripiprazole KANE2002 KANE2007B (perphenazine) CHAN2007B (risperidone) [4 weeks]
[4 weeks] [6 weeks] MCQUADE2004 (olanzapine) [26 weeks]*
KASPER2003 POTKIN2003A (risperidone) [4 weeks]
[52 weeks] ZIMBROFF2007 (ziprasidone) [4 weeks]
k=2 k=1 k=4
Clozapine Buchanan1998 [10 weeks] Claghorn1987 (chlorpromazine) Anand1998 (risperidone) [12 weeks]
Rosenheck1997 [52 weeks] [4–8 weeks] ATMACA2003 (olanzapine/
Tamminga1994 [52 weeks] Hong1997 (chlorpromazine) quetiapine/risperidone) [6 weeks]*
VOLAVKA2002 [14 weeks] [12 weeks] Beuzen1998 (olanzapine) [18 weeks]
Kane1988 (chlorpromazine) Bitter1999 (olanzapine) [18 weeks]
[6 weeks] Bondolfi1998 (risperidone) [8 weeks]
LIEBERMAN2003B [52 weeks]* Breier1999 (risperidone) [18 weeks]
Chowdhury1999 (risperidone) [16 weeks]
MELTZER2003A (olanzapine) [104 weeks]*
VOLAVKA2002 (olanzapine/risperidone)
[14 weeks]
k=4 k=4 k=9

Olanzapine Altamura1999 [14 weeks] Conley1998a (chlorpromazine) ATMACA2003 (quetiapine/risperidone)
Beasley1996a [6weeks] [8weeks] [6weeks]*
Beasley1997 [6 weeks] HGBL1997 (flupentixol) [4 weeks] Conley2001 (risperidone) [8 weeks]
Breier2000 [6 weeks] Jakovljevic1999 (fluphenazine) [6 DAVIDSON2007 (paliperidone) [6 weeks]
BUCHANAN2005 [16 weeks] weeks] Gureje1998 (risperidone) [30 weeks]
HGCJ1999 (HK) [14 weeks] Loza1999 (chlorpromazine) [6 weeks] Jones1998 (risperidone) [54 weeks]
HGCU1998 (Taiwan) [14 weeks] Naukkarinen1999/HGBJ KANE2007A (paliperidone) [6 weeks]
Jones1998 [54 weeks] (perphenazine) [26 weeks] KINON2006B (quetiapine) [26 weeks]
KONGSAKON2006 [24 weeks] Lecrubier1999 (amisulpride) [26 weeks]
LIEBERMAN2003A [24 weeks] MARDER2007 (paliperidone) [6 weeks]
LINDENMAYER2007 [12 weeks] MARTIN2002 (amisulpride) [24 weeks]
ROSENHECK2003] [52 weeks] MCEVOY2007A (quetiapine/ risperidone)
STUDY-S029 [52 weeks] [52 weeks]
Tollefson1997 [6 weeks] MCQUADE2004 (aripiprazole) [26 weeks]*
Tran1998a [52 weeks] RIEDEL2007B (quetiapine) [8 weeks]
Tran1998b [52 weeks] SIROTA2006 (quetiapine) [26 weeks]
Tran1998c [22–84 weeks] StudyS036 (risperidone) [6 weeks]
VOLAVKA2002 [14 weeks] Tran1997 (risperidone) [28 weeks]
VANNIMWEGEN2008 (risperidone)[6 weeks]
VOLAVKA2002 (risperidone) [14 weeks]
WAGNER2005 (amisulpride)[8 weeks]
k =18 k=5 k =19

Paliperidone - - DAVIDSON2007 (paliperidone)[6 weeks]
KANE2007A (paliperidone) [6 weeks]
MARDER2007 (paliperidone) [6 weeks]
k=3

Quetiapine Arvanitis1997 [6 weeks] CONLEY2005 (fluphenazine) ATMACA2003 (clozapine/
Emsley1999 [8 weeks] [12 weeks] olanzapine/risperidone) [6 weeks]*
Fleischhacker1996 [6 weeks] Link1994 (chlorpromazine) [6 weeks] CONLEY2005 (risperidone) [12 weeks]
Murasaki1999 [8 weeks] KINON2006B (olanzapine) [26 weeks]
Purdon2000 [26 weeks] RIEDEL2005 (risperidone) [12 weeks]
RIEDEL2007B (olanzapine) [8 weeks]
SIROTA2006 (olanzapine) [26 weeks]
ZHONG2006 (risperidone) [8 weeks]
k=5 k=2 k=7

Risperidone Blin1996 CONLEY2005 (fluphenazine) ATMACA2003 (olanzapine/quetiapine)
[4 weeks] [12 weeks] [6weeks]*
Ceskova1993 Hoyberg1993 (perphenazine) AZORIN2006 (sertindole)
[8 weeks] [8 weeks] [12weeks]
Chouinard1993 Huttunen1995 (zuclopenthixol) CHAN2007A (aripiprazole)
[8 weeks] [8 weeks] [4 weeks]
Claus1991 RUHRMANN2007 (flupentixol) Conley2001 (olanzapine)
[12 weeks] [25weeks] [8 weeks]
Csernansky1999/2000 CONLEY2005 (quetiapine)
[52 weeks] [12 weeks]
Emsley1995 Fleurot1997 (amisulpride)
[6 weeks] [8 weeks]
Heck2000 Gureje1998 (olanzapine)
Janicak1999 HWANG2003 (amisulpride)
[6 weeks] [6 weeks]
Jones1998 Jones1998 (olanzapine)
Kern1998 Klieser1996 (zotepine)
[8weeks] [4 weeks]
LEE2007 Lecrubier2000 (amisulpride)

Marder1994 MCEVOY2007A (olanzapine/quetiapine)
Mesotten1991 POTKIN2003A (aripiprazole)
[8 weeks] [4 weeks]
Min1993 RIEDEL2005 (quetiapine)
MOLLER2008 StudyS036 (olanzapine)
[8weeks] [6 weeks]
Peuskens1995 Tran1997 (olanzapine)
[8weeks] [28 weeks]
SCHOOLER2005 VANNIMWEGEN2008 (olanzapine)
SEE1999 [5 weeks] VOLAVKA2002 (clozapine/olanzapine) [14
ZHANG2001 [12 weeks] weeks]
VOLAVKA2002 [14 weeks] ZHONG2006 (quetiapine)
[8 weeks]
k = 20 k=4 k = 19
Sertindole Hale2000 [8 weeks] - AZORIN2006 (risperidone) [12 weeks]
Daniel1998 [52 weeks]*
k=2 k=1
Zotepine Barnas1987 [7 weeks] Cooper1999a (chlorpromazine) Klieser1996 (risperidone) [4 weeks]

Fleischhacker1989 [6 weeks] [8 weeks]
Klieser1996 [4 weeks]
KnollCTR (StudyZT4002)
[26 weeks]
Petit1996 [8 weeks]
k=5 k=1 k=1
Note.*Study did not meet the inclusion criteria for any other review reported in this chapter.

Pooling data from 138 evaluations of one antipsychotic versus another antipsychotic
did not reveal metabolic and neurological side effects that were inconsistent with
those reported in the SPC for each drug. Because most trials were of relatively short
duration and not designed to prospectively examine side effects, these trials provide
little insight into the longer-term adverse effects of treatment or whether there are
clinically significant differences between antipsychotic drugs.
10.8 EFFECTIVENESS OF ANTIPSYCHOTIC MEDICATION

10.8.1 Introduction
The RCT is widely recognised as the ‘gold standard’ for evaluating treatment
efficacy, but some methodological issues may compromise the generalisability of the
findings of research to the ordinary treatment setting. Nevertheless, it is still
recognised that the RCT is an indispensable first step in the evaluation of
interventions in mental health and provides the most valid method for determining
the impact of two contrasting treatment conditions (treatment efficacy), while
controlling for a wide range of participant factors including the effects of
spontaneous remission.
Once an approach has been demonstrated as efficacious under the stringent

conditions of an RCT, a next step is to examine its effectiveness in ordinary
treatment conditions, including large-scale effectiveness (pragmatic) trials (very few
of which were available when the 2002 guideline was developed).
In addition, the use of RCTs and other studies in the evaluation of interventions in
the treatment of schizophrenia is limited in many cases by the absence of important
outcome measures. For example, few trials report evidence on quality of life or
satisfaction with services, despite the fact that service users and carers view these
measures as very important. Effectiveness studies address this issue by focusing on
patient-important outcomes.
10.8.2 Effectiveness (pragmatic) trials

Given the large scope of the guideline update, the GDG decided to focus on
effectiveness trials that included a comparison between an SGA and an FGA. To
ensure that the evidence was from high-quality research and reduce the risk of bias,
studies were included only if they used a randomised design with an intention-to-
treat analysis and at least independent rater-blinding (that is, the clinicians doing the
assessment of outcome were independent and blind to treatment allocation). All
studies identified during the searches for other sections of this chapter were
considered for inclusion.
Two studies published since the 2002 guideline met the inclusion criteria for this
review. These were the CATIE study (Lieberman et al., 2005; Stroup et al., 2003),

funded by the National Institute of Mental Health, and the Cost Utility of the Latest
Antipsychotic Drugs in Schizophrenia Study (CUtLASS 1) (Jones et al., 2006; Lewis
et al., 2006b), funded by the NHS Research and Development Health Technology
Assessment Programme.
In the initial phase of CATIE (phase 1), which was conducted at 57 clinical sites in
the US, 1,493 participants with chronic schizophrenia were randomised (double-
blind) to one of four SGAs or an FGA (perphenazine) (see Table 113). Participants
with current tardive dyskinesia could enrol, but were not able to be randomised to
perphenazine. For the purposes of the 2009 guideline, the GDG focused on the
primary outcome (discontinuation of treatment for any reason), tolerability, and
both metabolic and neurological side effects. An evidence summary table for these
outcomes can be found in Appendix 23c (the section on effectiveness of
antipsychotic drugs).
In the initial phase of CUtLASS (Band 1), 227 participants with schizophrenia (or a
related disorder) were randomised to an FGA or SGA (the choice of individual drug
was made by the psychiatrist responsible for the care of the patient). The study was
conducted in 14 NHS trusts in England and was specifically designed to test
effectiveness in routine NHS practice. For the purposes of the 2009 guideline, the
GDG focused on the primary outcome (the Quality of Life Scale;(Heinrichs et al.,
1984)), tolerability, and neurological side effects. An evidence summary table for
these outcomes can be found in Appendix 23c (the section on effectiveness of
antipsychotic drugs).
Further analysis of cost effectiveness, including Band 2 of the CUtLASS trial can be
found in Section 10.9.
Table 113: Summary of study characteristics for the initial phases of CATIE and
CUtLASS
CATIE (Phase1) CUtLASS (Band1)

Total N 1,493 227
Diagnostic criteria DSM-IV DSM-IV
Intervention Number randomised (number Number randomised (most common at
that did not take drug): 52 weeks):
Olanzapine: 336 (6) FGA: 118 (26% were taking sulpiride)
Quetiapine: 337 (8) SGA: 109 (34% were taking olanzapine)
Risperidone: 341 (8)
Perphenazine: 261 (4)
Baseline severity– Olanzapine: 76.1 (18.2) FGA: 72.9 (17.2)
Mean PANSS (SD) Quetiapine: 75.7 (16.9) SGA: 71.3 (16.5)
Risperidone: 76.4 (16.6)
Perphenazine: 74.3 (18.1)

Selected inclusion Diagnosis of schizophrenia, no Diagnosis of schizophrenia (or
criteria history of serious adverse schizoaffective disorder or delusional
reactions to study medications, disorder), requiring change of current
not experiencing their first FGA or SGA treatment because of
episode, not treatment- inadequate clinical response or
resistant. intolerance, at least 1 month since the
first onset of positive psychotic
symptoms.
Setting Inpatient/outpatient Inpatient/outpatient
Duration of treatment Up to 18 months Up to 12 months
Medication dose Mean modal dose: Varied depending on drug taken
(mg/day) Olanzapine: 20.1 (n = 312)
Quetiapine: 534.4 (n = 309)
Risperidone: 3.9 (n = 305)
Perphenazine: 20.8 (n = 245)
Note. In the CATIE trial, after ~40% of participants were enrolled, ziprasidone was added as
treatment option and 185 participants were randomised to this arm. However, this drug is not
licensed in the UK and is therefore not included in this review.
a Thirty-three participants from one site were excluded from the analysis because of concerns regarding
the integrity of the data.

Two trials involving 1,720 participants failed to establish clinically significant
differences in effectiveness between the oral (non-clozapine) antipsychotic drugs
examined. Although both trials have limitations (for further information see
(Carpenter & Buchanan, 2008; Kasper & Winkler, 2006; Lieberman, 2006; Möller,
2008), it is clear that more effective medication is needed. Furthermore, neither study
included participants experiencing their first episode of schizophrenia or examined
depot/long- acting antipsychotic medication.
With regard to adverse effects of treatment, the diverse side effect profiles seen in
the efficacy trials reported elsewhere in this chapter were supported by CATIE and
CUtLASS and primarily confirmed differential metabolic effects. However, there
were no consistent clinically significant differences between antipsychotics in terms
of treatment-emergent EPS. It should be noted that the various FGAs tested (such as
perphenazine and sulpiride) were generally not high-potency antipsychotics and
were prescribed in standard doses. Further analyses of baseline data from CATIE
also confirm other reports that people with schizophrenia are undertreated for
metabolic disorders (Nasrallah et al., 2006).
10.9 HEALTH ECONOMICS

10.9.1 Systematic literature review
The systematic search of the economic literature, undertaken for the 2009 guideline,
identified 33 eligible studies on pharmacological treatments for people with
schizophrenia. Of these, one study assessed oral antipsychotic medications for initial
treatment of schizophrenia (Davies & Lewis, 2000); 15 studies examined oral drug

treatments for acute psychotic episodes (Alexeyeva et al., 2001; Almond &
O’Donnell, 2000; Bagnall et al., 2003; Beard et al., 2006; Bounthavong & Okamoto,
2007; Cummins et al., 1998; Edgell et al., 2000; Geitona et al., 2008; Hamilton et al.,
1999; Jerrell, 2002; Lecomte et al., 2000; Nicholls et al., 2003; Palmer et al., 2002;
Palmer et al., 1998; Rosenheck et al., 2003); eight studies assessed oral antipsychotic
medications aimed at promoting recovery (Davies et al., 1998; Ganguly et al., 2003;
Knapp et al., 2008; Launois et al., 1998; Oh et al., 2001; Rosenheck et al., 2006; Tunis
et al., 2006; Vera-Llonch et al., 2004); four studies examined pharmacological
treatments aiming at promoting recovery in people with schizophrenia whose illness
has not responded adequately to treatment (Davies et al., 2008; Lewis et al., 2006a;
Lewis et al., 2006b; Rosenheck et al., 1997; Tilden et al., 2002); and six studies
evaluated depot antipsychotic treatments (Chue et al., 2005; De Graeve et al., 2005;
Edwards et al., 2005; Heeg et al., 2008; Laux et al., 2005; Oh et al., 2001). Details on
the methods used for the systematic review of the economic literature in the 2009
guideline are described in Appendix 11; references to included and excluded studies
and evidence tables for all economic evaluations included in the systematic literature
review are provided in Appendix 25.
Initial treatment with antipsychotic medication

One study that assessed oral antipsychotics for the treatment of people with a first
episode of schizophrenia was included in the systematic economic literature review
(Davies & Lewis, 2000). The study, which was conducted in the UK, was a cost-
utility analysis based on a decision-analytic model in the form of a decision tree. The
antipsychotic treatments assessed were olanzapine, risperidone, chlorpromazine,
haloperidol and clozapine. All drugs, with the exception of clozapine, were assessed
as first, second, third or fourth lines of treatment, whereas clozapine was assessed as
a third or fourth line of treatment only. According to the model structure, people
switched to the next line of treatment when an antipsychotic was not acceptable to
them; treatment unacceptability was defined as treatment intolerance (development
of non-treatable or unacceptable side effects), inadequate response or non-
compliance. People who found treatment acceptable were transferred to
maintenance therapy. If they experienced a relapse during acceptable treatment over
the time frame of the analysis, they were treated with the same antipsychotic.
Acceptable side effects were treated without change in antipsychotic therapy. The
adverse events considered in the analysis were EPS (except tardive dyskinesia,
which was considered separately), tardive dyskinesia, neuroleptic malignant
syndrome, hepatic dysfunction and agranulocytosis. Clinical efficacy data were
derived from a systematic literature review and meta-analysis. The perspective of
the analysis was that of health and social care services including expenses of people
with schizophrenia. Resource use was based on published literature, other national
sources and further assumptions. Prices were taken from national sources. The time
horizon of the analysis was 3 years.
Results were reported separately for different scenarios regarding sequence of

antipsychotic treatments. Olanzapine and haloperidol were dominated by
chlorpromazine when used as any line of treatment. Risperidone was more effective

than chlorpromazine, but always at an additional cost, which reached £34,241 per
QALY when first-line treatment was assessed. Clozapine dominated olanzapine and
risperidone when used as third- or fourth-line treatment. It was shown to yield the
highest number of QALYs out of all antipsychotics included in the analysis. Its
incremental cost-effectiveness ratio (ICER) versus chlorpromazine was £35,689 and
£47,980 per QALY, when they were compared as third- and fourth-line treatments,
respectively.
The results of the analysis were statistically significant and indicated that olanzapine
and haloperidol were not cost-effective options compared with the other
antipsychotic drugs assessed for the treatment of people with a first episode of
schizophrenia. The authors concluded that clozapine (as third- or fourth-line
treatment) and risperidone might be more effective than chlorpromazine, but at a
higher cost. However, they recognised that because multiple comparisons of costs
and QALYs had been made, some statistically important differences might have
occurred by chance rather than reflected real differences. Moreover, they recognised
the limited availability of clinical data used in the model.
An additional limitation of the analysis was that efficacy data for each antipsychotic
medication were apparently derived from ‘naïve’ addition of data across relevant
treatment arms of all RCTs included in the systematic literature review. This method
treats the data as if they came from a single trial and practically breaks the
randomisation: data from treatment arms not directly relevant to the analysis are not
taken into account and between-trial variance is completely ignored (Glenny et al.,
2005). Glenny and colleagues argue that such a method of combining trial data is
liable to bias, highly unpredictable and also produces over-precise answers. They
conclude that results of such analysis are completely untrustworthy and, therefore,
naïve comparisons should never be made.
Furthermore, utility data used in the base-case analysis by Davies and Lewis (2000)
were based on published utility values of seven people with schizophrenia in
Canada (Glennie, 1997), which appeared to be favouring FGAs and clozapine.
Overall, the conclusions of this analysis should be interpreted with caution.
Oral antipsychotics in the treatment of the acute episode

The systematic review of the economic literature considered 15 studies evaluating
oral antipsychotic medications for the management of acute psychotic episodes
(Alexeyeva et al., 2001; Almond & O’Donnell, 2000; Bagnall et al., 2003; Beard et al.,
2006; Bounthavong & Okamoto, 2007; Cummins et al., 1998; Edgell et al., 2000;
Geitona et al., 2008; Hamilton et al., 1999; Jerrell, 2002; Lecomte et al., 2000; Nicholls
et al., 2003; Palmer et al., 2002; Palmer et al., 1998; Rosenheck et al., 2003) Of these,
four were conducted in the UK (Almond & O’Donnell, 2000; Bagnall et al., 2003;
Cummins et al., 1998; Nicholls et al., 2003)(and are described in more detail. Of the
remaining 11 studies, seven were conducted in the US (Alexeyeva et al., 2001;
Bounthavong & Okamoto, 2007; Edgell et al., 2000; Hamilton et al., 1999; Jerrell,
2002; Palmer et al., 1998; Rosenheck et al., 2003), one in Germany (Beard et al., 2006),

one in Belgium (Lecomte et al., 2000), one in Mexico (Palmer et al., 2002) and one in
Greece (Geitona et al., 2008). Bagnall et al. (2003), using the same economic model
structure as Davies and Lewis (2000), evaluated the cost effectiveness of SGAs for the
treatment of acute episodes in people with schizophrenia in the UK. Ten
antipsychotic medications were included in a cost-utility analysis: olanzapine,
risperidone, quetiapine, amisulpride, zotepine, sertindole, ziprasidone, clozapine,
chlorpromazine and haloperidol. Clinical data were based on a systematic literature
review and meta-analysis, and other published literature. The study adopted the
perspective of health and social care services. Resource use was based on published
literature and further assumptions. National unit costs were used. Outcomes were
expressed in QALYs. Utility values in the base-case analysis were also taken from
Glennie (1997). The time horizon of the analysis was 1 year.
Results were reported separately for first, second, third and fourth lines of treatment.
The authors performed comparisons between each SGA and the other medications.
Ziprasidone and amisulpride were associated with the highest costs and QALYs.
According to the authors, amisulpride was the most cost-effective SGA drug if
ziprasidone remained unlicensed. Amisulpride and ziprasidone were the most
effective and costliest drugs, followed by risperidone, which was both the third most
effective and costliest drug of those examined. Olanzapine was the least costly and
least effective antipsychotic. The authors suggested that sertindole, zotepine and
quetiapine were not superior to other SGAs in terms of cost effectiveness. However,
the cost and the effectiveness results were characterised by high uncertainty. In
addition, clinical data for haloperidol and chlorpromazine were taken from the
control arms of SGA trials because no systematic review of the literature was
undertaken for FGAs; this methodology may have introduced bias to the analysis. A
further limitation of the study was that analysis of efficacy data utilised the ‘naïve’
method for data pooling, as described earlier, and therefore the analysis is subject to
bias. For all of these reasons, no clear conclusions on the relative cost effectiveness of
SGAs can be drawn from this analysis, and this was also the authors’ conclusion.
Cummins et al. (1998) used the results of an RCT comparing olanzapine with
haloperidol for acute treatment of people with schizophrenia (TOLLEF- SON1997) to
inform a decision tree that was constructed to assess the relative cost effectiveness of
the two antipsychotic drugs in the UK. According to the model structure, people in
an acute episode were started on one of the two evaluated drugs and followed up
for 1 year. Those who did not respond to treatment, withdrew or relapsed following
any response had their medication switched to haloperidol (if they had been started
on olanzapine) or fluphenazine (if they had been started on haloperidol). The
perspective of the analysis was that of the NHS. Resource use was based on
published literature and further assumptions. Prices were taken from national
sources. Outcomes were expressed in QALYs. Utility values were estimated using
the index of health-related quality of life) (IHRQoL), a generic measure designed to
capture social, psychological and physical functioning.

Olanzapine was found to dominate haloperidol because it produced more QALYs
(0.833 versus 0.806) and resulted in lower costs (£26,200 versus £31,627). The results
were robust in a number of sensitivity analyses carried out. Limitations of the
analysis, as stated by the authors, were the weak evidence on longer-term effects of
antipsychotics, which led to a number of assumptions in the model, and the
simplicity of the model structure, which did not capture all events related to
treatment of acute episodes with antipsychotics.
Almond and O’Donnell (2000) conducted an economic analysis to compare the costs
and benefits associated with olanzapine, risperidone, and haloperidol in the
treatment of acute psychotic episodes in the UK. Analysis was based on decision-
analytic modelling. The economic model considered cycles of acute episodes,
remission and relapse over a period of 5 years. Efficacy data were taken from two
clinical trials (TOLLEFSON1997 and TRAN1997). The outcomes of the analysis were
the percentage of people with a Brief Psychiatric Rating Scale (BPRS) score below 18
and the percentage of people without relapse over the time frame of the analysis.
The study adopted the NHS perspective. Resource use estimates were based on
published literature and further assumptions. UK national prices were used.
Olanzapine was reported to be less costly than both risperidone and haloperidol
(costs of olanzapine, risperidone and haloperidol were £35,701, £36,590 and £36,653
respectively). In addition, olanzapine was found to be more effective (percentages of
people with a BPRS score below 18 over 5 years for olanzapine, risperidone and
haloperidol were 63.6%, 63.0%, and 52.2%, respectively; percentages of people
without relapse over 5 years were 31.2%, 29.3% and 18.2%, respectively). These
figures show that olanzapine and risperidone dominated haloperidol (olanzapine
was more effective at a lower cost; risperidone was more effective at a similar cost).
Olanzapine also dominated risperidone (it was slightly more effective at a lower
cost). Cost results were sensitive to daily dosages, relapse rates and dropout rates.
The authors reported as limitations of their analysis the assumptions needed to
estimate resource utilisation and the omission of some categories of cost, such as the
costs of monitoring drug therapy, owing to lack of relevant data.
Nicholls et al. (2003) performed a cost-minimisation analysis alongside an

international, multicentre clinical trial that compared amisulpride with risperidone
over a 6-month treatment period (LECRUBIER2000). The trial had demonstrated that
amisulpride and risperidone had similar effectiveness, as measured using the
Positive and Negative Syndrome Scale (PANSS), BPRS and Clinical Global
Impression (CGI) scale scores. The economic analysis, which adopted the
perspective of the NHS, utilised resource use estimates from the trial and UK unit
costs.
Amisulpride was found to be overall less costly than risperidone by £2,145, but the
result was not statistically significant (95% CI: −£5,379 to £1,089). The findings of the
study are not directly applicable to the UK setting, as resource use was based on
settings other than the UK, where clinical practice is likely to be different. For

example, part-time hospitalisations were recorded in some settings; the authors
stated that this type of care was not universally recognised in the NHS, and for this
reason respective UK unit costs were not available and needed to be based on
assumptions.
Of the further 11 studies included in the systematic review of the cost effectiveness
of oral antipsychotics in the management of acute psychotic episodes, nine involved
comparisons between olanzapine, risperidone and haloperidol. Relative cost
effectiveness between olanzapine and risperidone cannot be established with
certainty from the results of these studies: Beard et al. (2006) suggested that
olanzapine was dominant over risperidone because it was shown to be more
effective at a lower cost. The analysis, which was conducted from the perspective of
the German healthcare system, was based on decision-analytic modelling. Other
models of similar structure replicated this result in other countries: olanzapine
dominated risperidone in the US (Palmer et al., 1998) and in Mexico (Palmer et al.,
2002). On the other hand, the modelling studies by Bounthavong and Okamoto
(2007) in the US and (Lecomte et al., 2000) in Belgium indicated that risperidone
might be marginally dominant over olanzapine because it was associated with better
or similar outcomes at similar or slightly lower costs. Two economic analyses
conducted alongside clinical trials in the US (Edgell et al., 2000; Jerrell, 2002) were
also unable to draw certain conclusions: in both trials, olanzapine appeared to be less
costly than risperidone, but cost results were not statistically significant. In one of
the trials, olanzapine was associated with longer maintenance of response and lower
EPS rates (Edgell et al., 2000) but the other trial (Jerrell, 2002) failed to demonstrate a
superiority of olanzapine over risperidone in terms of clinical effectiveness.
With respect to the comparative cost effectiveness of olanzapine and haloperidol,

there was less variety in the study results: two modelling studies (Bounthavong &
Okamoto, 2007; Palmer et al., 1998) and one economic analysis undertaken along-
side a clinical trial (Hamilton et al., 1999) demonstrated that olanzapine dominated
haloperidol in the US because it was more effective at a lower cost. Another multi-
centre RCT conducted in the US (Rosenheck et al., 2003) showed that olanzapine had
similar effectiveness to haloperidol (measured by BPRS scores) and lower akathisia
rates. It was more expensive than haloperidol, but cost results were not statistically
significant. Finally, two modelling studies suggested that olanzapine was more
effective than haloperidol at an additional cost approximating £3 per day with
minimum symptoms and toxicity in Belgium (Lecomte et al., 2000) and £11,350 per
relapse avoided in Mexico (Palmer et al., 2002). Overall, these results suggest that
olanzapine may be more cost effective than haloperidol in the treatment of acute
episodes.
Two of the comparisons of risperidone versus haloperidol showed that risperidone

was the dominant option in the US (Bounthavong & Okamoto, 2007) and in Belgium
(Lecomte et al., 2000), while one economic model used to assessed the relative cost
effectiveness of the two antipsychotics in two different countries found risperidone
to be more effective than haloperidol at an additional cost that reached

$2,100/QALY in the US (Palmer et al., 1998) and about £13,900 per relapse avoided
in Mexico (Palmer et al., 2002). These findings suggest that risperidone may be more
cost effective than haloperidol.
Finally, of the remaining two studies included in the systematic economic literature
review of acute treatment for people with schizophrenia, the study conducted by
Alexeyeva and colleagues (2001) compared the cost effectiveness of olanzapine and
ziprasidone in the US; the study, which was based on decision-analytic modelling,
utilised published and unpublished clinical data and concluded that olanzapine
dominated ziprasidone because it was more effective at a similar total cost. The other
study (Geitona et al., 2008) assessed the cost effectiveness of paliperidone relative to
risperidone, olanzapine, quetiapine, aripiprazole and ziprasidone from the
perspective of the Greek healthcare system. The study, which was also based on
decision-analytic modelling, utilised efficacy data from selected placebo-controlled
trials and other published sources. Resource utilisation estimates were based on
expert opinion.
According to the authors’ conclusions, paliperidone was the most cost-effective drug
as it dominated all other treatment options assessed. This finding was reported to be
robust in sensitivity analysis. However, dominance of paliperidone over olanzapine
was only marginal (paliperidone resulted in 0.3 additional days free of symptoms
per year and an annual extra saving of €4 compared with olanzapine).
It must be noted that the results of most modelling studies were sensitive to changes
in response and dropout rates, drug acquisition costs, and hospitalisation rates for
an acute episode. Most of these studies did not maintain randomisation effects
because they used (and in some cases combined) efficacy data from arms of different
trials for each antipsychotic drug evaluated, using a ‘naïve’ method of pooling. The
impact of side effects on health related quality of life (HRQoL) was not explored in
the majority of them.
Promoting recovery in people with schizophrenia that is in remission-

pharmacological relapse prevention
Eight studies that were included in the systematic economic literature review
assessed oral antipsychotic medications for relapse prevention (Davies et al., 1998;
Ganguly et al., 2003; Knapp et al., 2008; Launois et al., 1998; Oh et al., 2001;
Rosenheck et al., 2006; Tunis et al., 2006; Vera-Llonch et al., 2004). None of the
studies was undertaken in the UK.
The most relevant study to the UK context was that by Knapp and colleagues (2008);
it evaluated the cost effectiveness of olanzapine versus a number of other
antipsychotic medications (including risperidone, quetiapine, amisulpride and
clozapine, as well as oral and depot FGAs) using clinical and resource use data from
a multicentre prospective observational study conducted in outpatient settings in ten
European countries. The analysis adopted the health service payer’s perspective;
costs were estimated by applying UK national unit cost data to recorded healthcare

resource use. Outcomes were expressed in QALYs, estimated by recording and
analysing participants’ EQ-5D scores and linking them to respective UK population
tariffs to determine utility values. The time horizon of the analysis was 12 months.
The study made separate comparisons of olanzapine with each of the other
antipsychotic medications considered; no direct comparisons were made between
the other antipsychotic medications. According to the performed comparisons,
olanzapine dominated quetiapine and amisulpride; it was more effective than
risperidone and clozapine at an additional cost reaching £5,156 and £775 per QALY,
respectively. Compared with oral and depot FGAs, olanzapine was more effective
and more costly, with an ICER of £15,696 and £23,331 per QALY respectively (2004
prices). However, FGAs were analysed together as a class, and no results from
comparisons between olanzapine and specific FGAs were reported. Probabilistic
sensitivity analysis conducted using bootstrap techniques revealed that the
probability of olanzapine being more cost effective than quetiapine was 100% at a
willingness-to-pay lower than £5,000/QALY; the probability of olanzapine being
cost effective when compared with risperidone and amisulpride was 100% at a
willingness-to-pay around £18,000/QALY; at a willingness-to-pay equalling £30,000
per QALY, the probability of olanzapine being more cost effective than clozapine,
oral FGAs and depot FGAs was 81%, 98% and 79% respectively.
The results of the analysis indicated that olanzapine had a high probability of being
cost effective relative to each of the other options assessed. However, no formal
incremental analysis across all comparators was performed, as all comparisons
involved olanzapine versus each of the other antipsychotics included in the analysis.
The study conclusions may have limited applicability in the UK because reported
healthcare resource use reflected average routine clinical practice in European
countries and only unit costs were directly relevant to the UK health service.
The rest of the economic studies on pharmacological relapse prevention mainly

included comparisons between olanzapine, risperidone and haloperidol. Two
modelling studies, one in Australia (Davies et al., 1998) and one in Canada (Oh et al.,
2001) concluded that risperidone was more cost effective than haloperidol because it
was more effective at a lower cost. One US modelling study reported that
risperidone was more effective and also more expensive than haloperidol (Ganguly
et al., 2003). The measure of outcome was the number of employable persons in each
arm of the analysis; employability was determined by a PANSS score reduction of at
least 20% from baseline and a WCST-Cat score of ≥3.5. The ICER of risperidone
versus haloperidol was estimated at $19,609 per employable person.
An economic analysis undertaken alongside an open-label trial in the US (Tunis et

al., 2006) showed that olanzapine was associated with better outcomes and lower
costs than risperidone in people with chronic schizophrenia, but results were
statistically insignificant. Another study based on mainly unpublished data and
employing Markov modelling techniques (Vera-Llonch et al., 2004) came to different
conclusions: according to this study, risperidone led to lower discontinuation rates,

had overall lower side effect rates and was less costly than olanzapine. A modelling
study carried out in France (Launois et al., 1998) reported that sertindole dominated
olanzapine and haloperidol; between olanzapine and haloperidol, the former was
the cost effective option. Overall, results of modelling studies were sensitive to
changes in response rates, compliance rates and hospital discharge rates.
Finally, Rosenheck and colleagues (2006) performed an economic analysis along-

side a large effectiveness trial in the US (CATIE, Lieberman et al., 2005). The study
compared olanzapine, quetiapine, risperidone, ziprasidone and perphenazine in
people with chronic schizophrenia. It was demonstrated that perphenazine
dominated all other antipsychotic medications, being significantly less costly than
the other antipsychotics but with similar effectiveness expressed in QALYs
(perphenazine was significantly more effective than risperidone at the 0.005 level in
intention-to-treat analysis). Differences in total healthcare costs were mainly caused
by differences in drug acquisition costs between perphenazine and the other
antipsychotic drugs considered.
Promoting recovery in people with schizophrenia whose illness has not

responded adequately to treatment (treatment resistance)
Four studies examining pharmacological treatments aiming at promoting recovery
in people with schizophrenia whose illness has not responded adequately to
treatment were included in the systematic review (Davies et al., 2008; Lewis et al.,
2006a; Lewis et al., 2006b; Rosenheck et al., 1997; Tilden et al., 2002).
Tilden and colleagues (2002) constructed a Markov model to assess the cost
effectiveness of quetiapine versus haloperidol in people with schizophrenia only
partially responsive to FGAs, from the perspective of the UK NHS. The model was
populated with clinical data taken from various sources: rates of response to
treatment were taken from a multicentre RCT, which compared two antipsychotics
in people with schizophrenia partially responsive to FGAs (EMSLEY1999). In this
study, response to treatment was defined as an improvement in PANSS total score of
at least 20% between the beginning and the end of the trial. Compliance rates in the
economic model were estimated by linking non-compliance with the presence of
EPS. Relapse rates were estimated by linking relapse with non-response to
treatment. Other clinical data were derived from published literature. Resource use
estimates were based on published studies and further assumptions; national unit
costs were used. The measure of outcome for the economic analysis was the average
number of relapses and the expected duration of time in response per person with
schizophrenia, over the time horizon of the analysis, which was 5 years. Quetiapine
was found to be more effective than haloperidol, at a slightly lower cost. Sensitivity
analysis revealed that cost results were sensitive to differences in response rates
between the two antipsychotic drugs, to the risk of relapse in non-responding and
non-compliant individuals, and to the proportion of people requiring hospitalisation
following relapse.

Rosenheck and colleagues (1997) assessed the cost effectiveness of clozapine relative
to haloperidol in people with schizophrenia refractory to treatment and a history of
high level use of inpatient services in the US, using a societal perspective. The
analysis was based on clinical and resource use evidence from a multicentre RCT
carried out in 15 Veterans Affairs medical centres. Clinical outcomes included
PANSS scores, Quality of Life Scale (QLS) scores, side effect rates and compliance
rates. Clozapine resulted in significantly lower mean PANSS scores, better
compliance rates and lower rates of EPS compared with haloperidol. The total
medical cost associated with clozapine was lower than the respective cost of
haloperidol, but the difference in costs was not statistically significant.
In addition to the above two studies, Lewis and colleagues (2006a) described two
effectiveness trials conducted in the UK that aimed at determining the clinical and
cost effectiveness of SGAs versus FGAs and clozapine versus SGAs in people with
schizophrenia responding inadequately to, or having unacceptable side effects from,
their current medication (CUtLASS, Bands 1 and 2). The studies would normally
have been excluded from the systematic review of the economic literature because
they treated SGAs and FGAs as classes of antipsychotic medications; no data relating
to specific antipsychotic drugs were reported. However, these studies were directly
relevant to the UK context and their findings could lead to useful conclusions
supporting formulation of guideline recommendations. Therefore, their methods
and economic findings are discussed in this section.
Both trials were conducted in adult mental health settings in 14 NHS trusts in
Greater Manchester, Nottingham and London. Participants in Band 1 (N = 227) were
randomised to either an SGA (olanzapine, risperidone, quetiapine or amisulpride) or
an FGA in oral or depot form. Participants in Band 2 (N = 136) were randomised to
either clozapine or one of the four SGAs named above. The primary clinical outcome
of the analyses was the QLS, with secondary outcomes PANSS scores, side effects
from medication and participant satisfaction. The measure of outcome in economic
analyses was the number of QALYs gained. QALYs were estimated by recording
and analysing participants’ EQ-5D scores and subsequently linking them to
respective UK population tariffs to determine utility values. Costs were estimated
from the perspective of health and social care services, and included medication,
hospital inpatient and outpatient services, primary and community care services and
social services. The time horizon of the analyses was 12 months.
According to the results for Band 1, FGAs dominated SGAs as they resulted in better
outcomes at a lower total cost, but the results were not statistically significant.
Bootstrap analysis of costs and QALYs, including imputed values for missing
observations and censored cases, demonstrated that FGAs resulted in 0.08 more
QALYs and net savings of £1,274 per person compared with SGAs (2001/02 prices).
In univariate sensitivity analyses, FGAs dominated SGAs or had an ICER lower than
£5,000 per QALY. Probabilistic sensitivity analysis (employing bootstrap techniques)
showed that at a zero willingness-to-pay, FGAs had a 65% probability of being cost
effective; this probability rose up to 91% at a willingness-to-pay equalling £50,000
per QALY. At a willingness-to-pay of £20,000 per QALY, the probability of FGAs

being more cost effective than SGAs was roughly 80%. The results of the economic
analysis indicate that FGAs are likely to be more cost effective than SGAs at the
NICE cost-effectiveness threshold of £20,000–£30,000 per QALY (NICE, 2008b).
According to the results for Band 2, clozapine resulted in a statistically significant

improvement in symptoms, but not in quality of life. Total costs associated with
clozapine were also significantly higher than respective costs of SGAs. Updated
bootstrap analysis of costs and QALYs showed that clozapine yielded 0.07 more
QALYs per person relative to SGAs, at an additional cost of £4,904 per person
(Davies et al., 2007). The ICER of clozapine versus SGAs was estimated at £33,240
per QALY (2005/06 prices). This value ranged from approximately £23,000 to
£70,000 per QALY in univariate sensitivity analyses. Probabilistic sensitivity analysis
showed that at a zero willingness-to-pay, clozapine had a 35% probability of being
cost effective compared with SGAs; this probability reached 50% at a willingness-to-
pay ranging between £30,000 and £35,000 per QALY. Results indicate that clozapine
is unlikely to be cost effective at the NICE cost-effectiveness threshold of £20,000 to
£30,000 per QALY (NICE, 2008b).
Analysis of costs in both trials revealed that the vast majority of costs (approximately
90% of total costs) were incurred by psychiatric hospital attendances; only 2 to 4% of
total costs constituted drug acquisition costs. Overall, there was great variance in the
use of health services and associated costs among study participants. The significant
difference in cost between clozapine and SGAs was caused by great difference in
psychiatric hospital costs between the two arms, possibly reflecting the licensing
requirement for inpatient admission for initiation of therapy with clozapine at the
time of the study. Currently, such requirements are no longer in place; therefore, at
present, the cost effectiveness of clozapine versus SGAs is likely to be higher than
demonstrated in the analysis.
Treatment with depot/long-acting injectable antipsychotic medication

The systematic review of the economic literature identified six studies assessing the
cost effectiveness of depot antipsychotic medications for people with schizophrenia
(Chue et al., 2005; De Graeve et al., 2005; Edwards et al., 2005; Heeg et al., 2008; Laux
et al., 2005; Oh et al., 2001). All studies were conducted outside the UK and
employed modelling techniques.
According to the results of these studies, long-acting risperidone was dominant over
haloperidol depot in Belgium (De Graeve et al., 2005), Germany (Laux et al., 2005),
Portugal (Heeg et al., 2008), Canada (Chue et al., 2005) and the US (Edwards et al.,
2005). Risperidone was dominant over olanzapine in Belgium (De Graeve et al.,
2005), Germany (Laux et al., 2005) and the US (Edwards et al., 2005). Risperidone
was dominant over oral risperidone in Portugal (Heeg et al., 2008), Canada (Chue et
al., 2005) and the US (Edwards et al., 2005). Finally, risperidone was also shown to
dominate quetiapine, ziprasidone and aripiprazole in the US (Edwards et al., 2005).
In all of the studies, the cost effectiveness of long-acting risperidone was largely
determined by its estimated higher compliance compared with oral antipsychotics.

However, in most studies, the methodology used to estimate compliance as well as
other clinical input parameters was not clearly described; a number of economic
models were populated with estimates based to a great extent on expert opinion.
Oh and colleagues (2001), using data from published meta-analyses and expert
opinion, reported that both haloperidol depot and fluphenazine depot were
dominated by oral risperidone in Canada. Although the methodology adopted was
clearly reported, the main limitation of this study was that randomisation effects
from clinical trials were not maintained because clinical input parameters were
estimated by pooling data from different clinical trials for each drug (‘naïve’ method
of synthesis).
Overall, the quality of evidence on depot antipsychotic medications was rather poor
and of limited applicability to the UK context, given that no study was conducted in
the UK.
The impact of compliance with antipsychotic treatment on healthcare

costs incurred by people with schizophrenia
The systematic search of economic literature identified a number of studies that
assessed the impact of non-adherence to antipsychotic medication on healthcare
costs incurred by people with schizophrenia. Although these studies did not
evaluate the cost effectiveness of specific pharmacological treatments and therefore
do not form part of the systematic review of economic evidence, they are described
in this section because they provide useful data on the association between
compliance, risk of relapse and subsequent healthcare costs. This information was
considered by the GDG at formulation of the guideline recommendations.
Knapp and colleagues (2004) analysed data from a national survey of psychiatric
morbidity among adults living in institutions in the UK, conducted in 1994.
Approximately 67% of the population surveyed had a diagnosis of schizophrenia.
According to the data analysis, non-adherence was one of the most significant
factors that increased health and social care costs. Non-adherence predicted an
excess annual cost reaching £2,500 per person for inpatient services and another
£2,500 for other health and social care services, such as outpatient and day care,
contacts with community psychiatric nurses, occupational therapists and social
workers, and sheltered employment (2001 prices).
A modelling exercise that simulated the treated course of schizophrenia assessed the
impact of compliance on health benefits and healthcare costs in people with
schizophrenia in the UK over a period of 5 years (Heeg et al., 2005). The study
considered people experiencing a second or third episode of schizophrenia and took
into account factors such as gender, disease severity, potential risk of harm to self
and society, and social and environmental factors. Other factors, such as number of
psychiatric consultations, presence of psychotic episodes, symptoms and side effects,
were also incorporated into the model structure. People with a first episode of
schizophrenia were excluded from the analysis. The analysis demonstrated that a
20% increase in compliance with antipsychotic treatment resulted in cost savings of

£16,000 and in prevention of 0.55 psychotic episodes per person with schizophrenia
over 5 years. Cost savings were almost exclusively attributed to the great reduction
in hospitalisation costs following improved compliance. Higher levels of compliance
were also associated with increased time between relapses, decreased symptom
severity and improved ability of people to take care of themselves.
With regard to people experiencing a first episode of schizophrenia, Robinson and

colleagues (1999) assessed the rates of relapse following response to antipsychotic
treatment in 104 people with a first episode of schizophrenia or schizoaffective
disorder. The authors reported that, after initial recovery, the cumulative first-
relapse rate was 82% over 5 years. Discontinuation of pharmacological treatment
increased the risk of relapse by almost five times. The authors concluded that the
risk of relapse within 5 years of recovery from a first episode of schizophrenia or
schizoaffective disorder was high, but could be diminished with maintenance
antipsychotic drug therapy. Although the study did not assess the costs associated
with non-compliance, its results indicate that compliance with treatment can reduce
healthcare costs considerably by reducing rates of relapse (relapse can lead to high
hospitalisation costs).
Finally, two published reviews examined the impact of compliance with

antipsychotic therapy on healthcare costs incurred by people with schizophrenia
(Sun et al., 2007; Thieda et al., 2003). The reviews analysed data from 21 studies in
total and concluded that antipsychotic non-adherence led to an increase in relapse
and, subsequently, hospitalisation rates and hospitalisation costs.
Summary of findings and conclusions from systematic economic literature

review
The economic literature review included 31 economic evaluations of specific
antipsychotic treatments for the management of people with schizophrenia, plus two
effectiveness trials conducted in the UK, which assessed antipsychotic medications
grouped in classes. Twenty-two studies were based on decision-analytic modelling
and were characterised by varying quality with respect to sources of clinical and
utility data and methods of evidence synthesis. Clinical data were derived from a
variety of sources, ranging from published meta-analyses and RCTs to unpublished
trials and expert opinion. Even when data were taken from meta-analyses of trial
data, the effects of randomisation were not retained, because data were simply
pooled (by using weighted mean values) from the respective trials evaluating the
drug under assessment. This ‘naïve’ method is likely to have introduced strong bias
in the analyses, and therefore is inappropriate for evidence synthesis of trial data
(Glenny et al., 2005). The impact of side effects on the HRQoL was explored in few
studies, and even in these cases it was the decrement in HRQoL owing to the
presence of EPS that was mostly considered. The impact of other side effects on
HRQoL was not explored. The majority of the studies were funded by industry,
which may have resulted in additional bias.

The included studies reported a variety of findings. The results of modelling
exercises were sensitive, as expected, to a number of parameters, such as response
and dropout rates, as well as rates and/or length of hospitalisation. Most of the cost
results derived from clinical studies were statistically insignificant. With the
exception of a few studies, the majority of economic evaluations included a very
limited number of antipsychotic medications for the treatment of people in
schizophrenia, mainly olanzapine, risperidone and haloperidol; however, a wider
variety of antipsychotic medications has been shown to be clinically effective and is
available in the market. Results of comparisons between the three most examined
drugs were in some cases contradictory. Nevertheless, overall findings of the
systematic review seem to suggest that olanzapine and risperidone may be more cost
effective than haloperidol. Similarly, there is evidence that long-acting risperidone
may lead to substantial cost- savings and higher clinical benefits compared with oral
forms of antipsychotic medication because of higher levels of adherence
characterising long-acting injectable forms. However, evidence on long-acting
injectable forms comes from non-UK modelling studies that are characterised by
unclear methods in estimating a number of crucial input parameters (such as levels
of adherence).
The results of non-UK studies are not directly applicable to the UK context and
therefore, although they may be indicative of trends in relative cost effectiveness of
different antipsychotic drugs worldwide, they should not be used exclusively to
inform decisions in the UK context. On the other hand, the results of UK studies
were characterised by high uncertainty and several important limitations.
The results of the economic analyses alongside effectiveness trials in the UK (Davies
et al., 2008; Lewis et al., 2006b) suggest that hospitalisation costs are the drivers of
total costs associated with treatment of people with schizophrenia. Drug acquisition
costs are only a small part of total costs, and are unlikely to affect significantly the
cost effectiveness of antipsychotic medications. It could be hypothesised that in the
short term and for people with schizophrenia treated as inpatients (for example,
during an acute episode), there are no big differences in total costs between
antipsychotic medications, unless there are differences in the length of hospital stays.
It might be reasonable to argue that antipsychotic drugs that reduce the rate and
length of hospital admissions (for example drugs that reduce the rate of future
relapses and/or the length of acute episodes) are cost-saving options in the long
term, despite potentially high acquisition costs. A related factor affecting the
magnitude of healthcare costs and subsequently the cost effectiveness of
antipsychotic medications is the level of adherence: according to published evidence,
high levels of adherence to antipsychotic treatment can greatly reduce the risk of
relapse and subsequent hospitalisation costs.
Details of the methods and the results of all economic evaluations described in this
section are provided in Appendix 25.

10.9.2 Economic modelling
A decision-analytic model was developed to assess the relative cost effectiveness of
antipsychotic medications aimed at promoting recovery (preventing relapse) in
people with schizophrenia in remission. The rationale for economic modelling, the
methodology adopted, the results and the conclusions from this economic analysis
are described in detail in Chapter 11. This section provides a summary of the
methods employed and the results of the economic analysis.
Overview of methods
A Markov model was constructed to evaluate the relative cost effectiveness of a
number of oral antipsychotic medications over two different time horizons, that is,
10 years and over a lifetime. The antipsychotic drugs assessed were olanzapine,
amisulpride, zotepine, aripiprazole, paliperidone, risperidone and haloperidol. The
choice of drugs was based on the availability of relapse prevention data identified in
clinical evidence review (see Section 10.4). The study population consisted of people
with schizophrenia in remission. The model structure considered events such as
relapse, discontinuation of treatment because of intolerable side effects and
switching to another antipsychotic drug, discontinuation of treatment because of
other reasons and moving to no treatment, development of side effects such as acute
EPS, weight gain, diabetes and glucose intolerance, complications related to diabetes
and death. Clinical data were derived from studies included in the guideline
systematic review of clinical evidence and other published literature. Where
appropriate, clinical data were analysed using mixed treatment comparison or
standard meta-analytic techniques. The measure of outcome in the economic
analysis was the number of QALYs gained. The perspective of the analysis was that
of health and personal social care services. Resource use was based on published
literature, national statistics and, where evidence was lacking, the GDG expert
opinion. National UK unit costs were used. The cost year was 2007. Two methods
were employed for the analysis of input parameter data and presentation of the
results. First, a deterministic analysis was undertaken, where data were analysed as
point estimates and results were presented in the form of ICERs following the
principles of incremental analysis. A probabilistic analysis was subsequently
performed in which most of the model input parameters were assigned probability
distributions. This approach allowed more comprehensive consideration of the
uncertainty characterising the input parameters and captured the non-linearity
characterising the economic model structure. Results of probabilistic analysis were
summarised in the form of cost effectiveness acceptability curves, which express the
probability of each intervention being cost effective at various levels of willingness-
to-pay per QALY gained (that is, at various cost- effectiveness thresholds).
Overview of results
Results of deterministic analysis demonstrated that zotepine dominated all other
treatment options, as it was less costly and resulted in a higher number of QALYs,
both at 10 years and over a lifetime of antipsychotic medication use. After zotepine,
olanzapine and paliperidone appeared to be the second and third most cost-effective

drugs respectively, in both time horizons of 10 years and over a lifetime.
Paliperidone and olanzapine dominated all other drugs (except zotepine) at 10 years;
the ICER of paliperidone versus olanzapine was approximately £150,000/QALY.
Over a lifetime, olanzapine was shown to be the least effective and least costly
intervention among those examined, but according to incremental analysis it was
still ranked as the second most cost-effective option following zotepine, using a cost-
effectiveness threshold of £20,000/QALY (note that adopting a threshold of
£30,000/QALY would result in paliperidone being ranked the second most cost-
effective option and olanzapine third, as the ICER of paliperidone versus olanzapine
was just above the £20,000/QALY threshold, at £20,872/QALY). According to
sensitivity analysis, results were highly sensitive to the probability of relapse
attached to each antipsychotic drug, but were not driven by the estimated
probabilities of developing each of the side effects considered in the analysis.
Probabilistic analysis revealed that zotepine had the highest probability of being the
most cost-effective option among those assessed, but this probability was rather low,
roughly 27 to 30%, reflecting the uncertainty characterising the results of the
analysis. This probability was practically independent of the cost-effectiveness
threshold and the time horizon examined. The other antipsychotic medications had
probabilities of being cost effective that ranged from approximately 5% (haloperidol)
to 16% (paliperidone). Again, these probabilities were rather unaffected by different
levels of willingness-to-pay and consideration of different time horizons.
The results of the economic analysis are characterised by substantial levels of

uncertainty as illustrated in probabilistic analysis, indicating that no antipsychotic
medication can be considered clearly cost effective compared with the other options
included in the assessment. Moreover, it needs to be emphasised that the evidence
base for the economic analysis was in some cases limited because clinical data in the
area of relapse prevention for three medications (zotepine, paliperidone and
aripiprazole) came from three single placebo-controlled trials.

In the 2002 guideline (which incorporated the recommendations from the NICE
technology appraisal of SGAs [NICE, 2002]), SGAs were recommended in some
situations as first-line treatment, primarily because they were thought to carry a
lower potential risk of EPS. However, evidence from the updated systematic reviews
of clinical evidence presented in this chapter, particularly with regard to other
adverse effects such as metabolic disturbance, and together with new evidence from
effectiveness (pragmatic) trials, suggest that choosing the most appropriate drug and
formulation for an individual may be more important than the drug group.
Moreover, design problems in the individual trials continue to make interpretation

of the clinical evidence difficult. Such problems include: (a) high attrition from one
or both treatment arms in many studies; (b) differences between treatment arms in
terms of medication dose; (c) small numbers of studies reporting the same outcomes
for some drugs.

For people with schizophrenia whose illness has not responded adequately to
antipsychotic medication, clozapine continues to have the most robust evidence for
efficacy. In addition, evidence from the effectiveness studies (CATIE, Phase 2;
CUtLASS, Band 2) suggests that in people who have shown a poor response to non-
clozapine SGAs, there is an advantage in switching to clozapine rather than another
SGA. Nevertheless, even with optimum clozapine treatment it seems that only 30 to
60% of treatment- resistant illnesses will respond satisfactorily (Chakos et al., 2001;
Iqbal et al., 2003).
The systematic review of the economic literature identified a number of studies of

varying quality and relevance to the UK setting. Results were characterised, in most
cases, by high uncertainty. The majority of studies assessed the relative cost
effectiveness between olanzapine, risperidone and haloperidol. Although study
findings are not consistent, they seem to indicate that, overall, olanzapine and
risperidone might be more cost effective than haloperidol.
In the area of antipsychotic treatment for first episode or early schizophrenia, the
economic evidence is limited and characterised by important limitations, and
therefore no safe conclusions on the relative cost effectiveness of antipsychotic
medications can be drawn.
The amount of economic evidence is substantially higher in the area of

pharmacological treatment for people with an acute exacerbation or recurrence of
schizophrenia. However, the number of evaluated drugs is very limited and does
not cover the whole range of drugs licensed for treatment of people with
schizophrenia in the UK. In addition, existing studies are characterised by a number
of limitations and, in many cases, by contradictory results. Available evidence
indicates that olanzapine and risperidone may be more cost-effective options than
haloperidol for acute exacerbation or recurrence of schizophrenia.
The economic literature in the area of relapse prevention is characterised by similar

methodological limitations and also by the limited number of drugs assessed.
Olanzapine and risperidone have been suggested to be more cost effective than
haloperidol in preventing relapse, but these conclusions are based on results from
analyses conducted outside the UK. On the other hand, evidence from CATIE
suggests that perphenazine may be more cost effective than a number of SGAs (that
is, olanzapine, quetiapine, risperidone and ziprasidone) in the US.
For people with schizophrenia whose illness has not responded adequately to
treatment, sparse data on the cost effectiveness of specific antipsychotic medications
are available. Evidence from CUtLASS, although not providing data on the cost
effectiveness of individual drugs, provides useful insight into the factors that affect
total costs incurred by people with schizophrenia. According to economic findings
from CUtLASS, psychiatric inpatient care costs are the drivers of total healthcare

costs incurred by people with schizophrenia, with drug acquisition costs being only
a small fraction of total costs.
CUtLASS Band 2 found that clozapine was more effective than SGAs in the
treatment of people with inadequate response to, or unacceptable side effects from,
current medication, but at a higher cost that reached £33,000/QALY (ranging from
£23,000 to £70,000/QALY in univariate sensitivity analysis). It was suggested that
the significant difference in cost between clozapine and SGAs might have been
caused by a great difference in psychiatric hospital costs between clozapine and
SGAs, possibly reflecting the licensing requirement for inpatient admission for
initiation of therapy with clozapine at the time of the study. Currently, clozapine can
be initiated in an outpatient setting; therefore, the current cost effectiveness of
clozapine versus SGAs for people with inadequate response to treatment or
unacceptable side effects is likely to be higher than was estimated when CUtLASS
Band 2 was conducted.
Regarding depot/long-acting injectable antipsychotic medication, there is evidence

that long-acting risperidone may lead to substantial cost savings and greater clinical
benefits compared with oral forms of antipsychotic medication because of higher
levels of adherence characterising long-acting injectable forms. However, this
evidence comes from non-UK modelling studies that are characterised by unclear
methods in estimating a number of crucial input parameters.
The economic analysis undertaken for this guideline estimated the cost effectiveness
of oral antipsychotic medications for relapse prevention in people with
schizophrenia. The results of the analysis suggest that zotepine is potentially the
most cost-effective oral antipsychotic drug included in the model. However, results
were characterised by high uncertainty and probabilistic analysis showed that no
antipsychotic medication could be considered to be clearly cost effective compared
with the other treatment options assessed: according to results of probabilistic
analysis, the probability of each drug being cost effective ranged from roughly 5%
(haloperidol) to about 27 to 30% (zotepine), and was independent of the cost
effectiveness threshold used and the time horizon of the analysis (that is, 10 years or
a lifetime). The probability of 27 to 30% assigned to zotepine, although indicative, is
rather low and inadequate to be able to come to a safe conclusion regarding
zotepine’s superiority over the other antipsychotics assessed in terms of cost
effectiveness. Moreover, clinical data for zotepine in the area of relapse prevention
were exclusively derived from one small placebo-controlled RCT. Similarly, clinical
data for paliperidone and aripiprazole were taken from two placebo-controlled
trials. It must be noted that the economic analysis did not examine the cost
effectiveness of quetiapine and any FGAs apart from haloperidol, owing to lack of
respective clinical data in the area of relapse prevention.
An interesting finding of the economic analysis was that drug acquisition costs did
not affect the cost effectiveness of antipsychotic medications: in fact haloperidol,
which has the lowest price in the UK among those assessed, appeared to have the

lowest probability (about 5%) of being cost effective at any level of willingness-to-
pay. On the other hand, zotepine, which had the lowest average relapse rate across
all evaluated treatments, dominated all other options in deterministic analysis and
demonstrated the highest probability of being cost effective in probabilistic analysis;
this finding together with results of sensitivity analysis indicate that the effectiveness
of an antipsychotic drug in preventing relapse is the key determinant of its relative
cost effectiveness, apparently because relapse prevention, besides clinical
improvement, leads to a substantial reduction in hospitalisation rates and respective
costs.
Hospitalisation costs have been shown to drive healthcare costs incurred by people
with schizophrenia, both in published evidence and in the economic analysis carried
out for this guideline. It might be reasonable to argue that antipsychotic drugs that
reduce the rate and length of hospital admissions (for example, drugs that reduce the
rate of future relapses and/or the length of acute episodes) are cost-saving options in
the long term, despite potentially high acquisition costs. This hypothesis is
supported by published evidence, which shows that increased adherence to
antipsychotic treatment is associated with a significant decrease in healthcare costs
incurred by people with schizophrenia through a reduction in the risk of relapse and
subsequent need for hospitalisation.
The GDG considered all clinical and economic evidence summarised in this section
to formulate recommendations. In therapeutic areas where clinical and/or economic
evidence on specific antipsychotic medications was lacking, as in the case of
quetiapine and FGAs other than haloperidol in the area of relapse prevention, the
GDG made judgements on the clinical and cost effectiveness of antipsychotic
medication by extrapolating existing evidence and conclusions from other
therapeutic areas.
Taking into account the findings from the systematic reviews of both the clinical and
health economic literature, and the uncertainty characterising the results of economic
modelling undertaken for this guideline, the evidence does not allow for any general
recommendation for one antipsychotic to be preferred over another, but the evidence
does support a specific recommendation for clozapine for people whose illness does
not respond adequately to other antipsychotic medication.
Finally, the GDG noted that the following are the key points to be considered before
initiating an antipsychotic medication in an acute episode of schizophrenia. First,
there may be some lack of insight into the presence of a mental illness and the
relevance of drug treatment. Careful explanation is needed regarding the rationale
for antipsychotic medications and their modes of action. People with schizophrenia
will usually accept that they have been stressed, experiencing insomnia and not
eating well, so the acceptance of a tranquillising medication to help reduce stress and
improve sleep and appetite might be acceptable. It can also be explained, if the
patient is insightful enough, that the medication is antipsychotic and can help reduce
the severity of distressing hallucinations, delusions and thought disorder.

Second, medication should always be started at a low dose if possible, after a full
discussion of the possible side effects. Starting at a low dose allows monitoring for
the early emergence of side effects, such as EPS, weight gain or insomnia. The dose
can then be titrated upwards within the BNF treatment range. Although
polypharmacy with antipsychotic medications is not recommended, it is equally
important not to under treat the acute psychotic episode.
Third, people with schizophrenia should be consulted on their preference for a more
or less sedative medication option. Medication is ideally started following a period
of antipsychotic-free assessment within an acute ward setting or under the
supervision of a crisis home treatment team, early intervention in psychosis team or
assertive outreach team.**2009**
Following the publication of Psychosis and Schizophrenia in Children and Young People,
the GDG for the 2014 guideline took the view that the recommendations should be
consistent where appropriate, including changing the population from ‘people with
schizophrenia’ to ‘people with psychosis and schizophrenia’. The GDG also wished
to make it explicit that the options for first episode psychosis and for an acute
exacerbation or recurrence of psychosis or schizophrenia should be oral
antipsychotic medication combined with psychological interventions (individual
CBT and family intervention). This does not constitute a change to the meaning or
content of the original recommendations about antipsychotics, and it continues to
reflect the evidence. Rather, it clarifies what was implicit in the 2009 guideline, that
all people with psychosis and schizophrenia should be offered antipsychotic
medication together with a psychological intervention for both a first episode and
for subsequent exacerbations.
The GDG also considered the physical health of the service user and the effects of
antipsychotic medication on mortality and morbidity. The GDG suggested that
when antipsychotic medication is initiated for the first time as well as thought-out
treatment with antipsychotic medication, it is important that the physical health of
the service user is assessed and monitored. The GDG thought that was well as
collecting data of baseline measurements of weight and waist circumference, and
possible cardiovascular risks (using blood and pulse pressure), indicators of
possibility future weight gain, for example, levels of physical activity, eating habits,
and any current or emerging physical movement restrictions, should also be
investigated.**2009**

10.11 RECOMMENDATIONS
Treatment for first episode psychosis
conjunction with
10.11.1.2 The choice of antipsychotic medication should be made by the service

user and healthcare professional together, taking into account the views of
the carer if the service user agrees. Provide information and discuss the
likely benefits and possible side effects of each drug, including:
• metabolic (including weight gain and diabetes)
• extrapyramidal (including akathisia, dyskinesia and dystonia)
• cardiovascular (including prolonging the QT interval)
• hormonal (including increasing plasma prolactin)
• other (including unpleasant subjective experiences). [2009; amended 2014]
How to use oral antipsychotics

10.11.1.3 Before starting antipsychotic medication, undertake and record the
following baseline investigations:
• weight (plotted on a chart)
• waist circumference
• pulse and blood pressure
• fasting blood glucose, glycosylated haemoglobin (HbA1c), blood lipid profile
and prolactin levels
• assessment of any movement disorders
• assessment of nutritional status, diet and level of physical activity. [new 2014]
10.11.1.4 Before starting antipsychotic medication, offer the person with

psychosis or schizophrenia an electrocardiogram (ECG) if:
• specified in the summary of product characteristics (SPC)
• a physical examination has identified specific cardiovascular risk (such as
diagnosis of high blood pressure)
• there is a personal history of cardiovascular disease or
• the service user is being admitted as an inpatient. [2009]
10.11.1.5 Treatment with antipsychotic medication should be considered an

explicit individual therapeutic trial. Include the following:
• Discuss and record the side effects that the person is most willing to tolerate.
• Record the indications and expected benefits and risks of oral antipsychotic
medication, and the expected time for a change in symptoms and appearance
of side effects.
• At the start of treatment give a dose at the lower end of the licensed range and
slowly titrate upwards within the dose range given in the British national
formulary (BNF) or SPC.
• Justify and record reasons for dosages outside the range given in the BNF or
SPC.
• Record the rationale for continuing, changing or stopping medication, and the
effects of such changes.
• Carry out a trial of the medication at optimum dosage for 4–6 weeks. [2009;
amended 2014]
10.11.1.6 Monitor and record the following regularly and systematically

throughout treatment, but especially during titration:
• response to treatment, including changes in symptoms and
behaviour
• side effects of treatment, taking into account overlap between
certain side effects and clinical features of schizophrenia (for
example, the overlap between akathisia and agitation or anxiety)
and impact on functioning
• the emergence of movement disorders
• weight, weekly for the first 6 weeks, then at 12 weeks, at 1 year and
then annually (plotted on a chart)
• waist circumference annually (plotted on a chart)
• pulse and blood pressure at 12 weeks, at 1 year and then annually
• fasting blood glucose, HbA1c and blood lipid levels at 12 weeks, at
1 year and then annually
• adherence
• overall physical health. [new 2014]
10.11.1.7 The secondary care team should maintain responsibility for monitoring
service users’ physical health and the effects of antipsychotic medication for
at least the first 12 months or until the person’s condition has stabilised,
whichever is longer. Thereafter, the responsibility for this monitoring may
be transferred to primary care under shared care arrangements. [new 2014]
10.11.1.8 Discuss any non-prescribed therapies the service user wishes to use
(including complementary therapies) with the service user, and carer if
appropriate. Discuss the safety and efficacy of the therapies, and possible
interference with the therapeutic effects of prescribed medication and
psychological treatments. [2009]
10.11.1.9 Discuss the use of alcohol, tobacco, prescription and non-prescription
medication and illicit drugs with the service user, and carer if appropriate.
Discuss their possible interference with the therapeutic effects of prescribed
medication and psychological treatments. [2009]

10.11.1.10 ‘As required’ (p.r.n.) prescriptions of antipsychotic medication should
be made as described in recommendation 10.11.1.5. Review clinical
indications, frequency of administration, therapeutic benefits and side
effects each week or as appropriate. Check whether ‘p.r.n.’ prescriptions
have led to a dosage above the maximum specified in the BNF or SPC. [2009]
10.11.1.11 Do not use a loading dose of antipsychotic medication (often referred
to as ‘rapid neuroleptisation’). [2009]
10.11.1.12 Do not initiate regular combined antipsychotic medication, except for
short periods (for example, when changing medication). [2009]
10.11.1.13 If prescribing chlorpromazine, warn of its potential to cause skin
photosensitivity. Advise using sunscreen if necessary. [2009]
Treatment of acute episode

conjunction with
• psychological interventions (family intervention and individual CBT ,

schizophrenia, offer oral antipsychotic medication or review existing
medication. The choice of drug should be influenced by the same criteria
recommended for starting treatment (see recommendations 10.11.1.2–
10.11.1.13). Take into account the clinical response and side effects of the
service user’s current and previous medication. [2009; amended 2014]
Behaviour that challenges

10.11.1.16 Occasionally people with psychosis or schizophrenia pose an
immediate risk to themselves or others during an acute episode and may need
rapid tranquillisation. The management of immediate risk should follow the
relevant NICE guidelines (see recommendations 10.11.1.17 and 10.11.1.20).
[2009]
10.11.1.17 Follow the recommendations in Violence (NICE clinical guideline 25)
when facing imminent violence or when considering rapid tranquillisation.
[2009]
10.11.1.18 After rapid tranquillisation, offer the person with psychosis or
schizophrenia the opportunity to discuss their experiences. Provide them with
a clear explanation of the decision to use urgent sedation. Record this in their
notes. [2009]
10.11.1.19 Ensure that the person with psychosis or schizophrenia has the
opportunity to write an account of their experience of rapid tranquillisation in
their notes. [2009]

10.11.1.20 Follow the recommendations in Self-harm (NICE clinical guideline 16)
when managing acts of self-harm in people with psychosis or schizophrenia.
[2009]
Early post-acute period

10.11.1.21 Inform the service user that there is a high risk of relapse if they stop
medication in the next 1–2 years. [2009]
10.11.1.22 If withdrawing antipsychotic medication, undertake gradually and
monitor regularly for signs and symptoms of relapse. [2009]
10.11.1.23 After withdrawal from antipsychotic medication, continue monitoring
for signs and symptoms of relapse for at least 2 years. [2009]
Promoting recovery
10.11.1.24 Review antipsychotic medication annually, including observed benefits
and any side effects. [new 2014].
10.11.1.25 The choice of drug should be influenced by the same criteria
recommended for starting treatment (see recommendations 10.11.1.2-
10.11.1.13). [2009]
10.11.1.26 Do not use targeted, intermittent dosage maintenance strategies 40
routinely. However, consider them for people with psychosis or schizophrenia
who are unwilling to accept a continuous maintenance regimen or if there is
another contraindication to maintenance therapy, such as side-effect
sensitivity. [2009]
10.11.1.27 Consider offering depot /long-acting injectable antipsychotic
medication to people with psychosis or schizophrenia:
• who would prefer such treatment after an acute episode
• where avoiding covert non-adherence (either intentional or
unintentional) to antipsychotic medication is a clinical priority
within the treatment plan. [2009]
Using depot/long-acting injectable antipsychotic medication

10.11.1.28 When initiating depot/long-acting injectable antipsychotic medication:
• take into account the service user’s preferences and attitudes
towards the mode of administration (regular intramuscular
injections) and organisational procedures (for example, home visits
and location of clinics)
• take into account the same criteria recommended for the use of oral
antipsychotic medication (see recommendations 10.11.1.2–
40 Defined as the use of antipsychotic medication only during periods of incipient relapse or symptom
exacerbation rather than continuously.

10.11.1.13), particularly in relation to the risks and benefits of the
drug regimen
• initially use a small test dose as set out in the BNF or SPC. [2009]
Interventions for people whose illness has not responded adequately to

treatment
10.11.1.29 For people with schizophrenia whose illness has not responded
adequately to pharmacological or psychological treatment:
• Review the diagnosis.
• Establish that there has been adherence to antipsychotic medication,
prescribed at an adequate dose and for the correct duration.
• Review engagement with and use of psychological treatments and ensure that
these have been offered according to this guideline. If family intervention has
been undertaken suggest CBT; if CBT has been undertaken suggest family
intervention for people in close contact with their families.
• Consider other causes of non-response, such as comorbid substance misuse
(including alcohol), the concurrent use of other prescribed medication or
physical illness. [2009]
10.11.1.30 Offer clozapine to people with schizophrenia whose illness has not
responded adequately to treatment despite the sequential use of adequate
doses of at least 2 different antipsychotic drugs. At least 1 of the drugs should
be a non-clozapine second-generation antipsychotic. [2009]
10.11.1.31 For people with schizophrenia whose illness has not responded
adequately to clozapine at an optimised dose, healthcare professionals should
consider recommendation 10.11.1.29 (including measuring therapeutic drug
levels) before adding a second antipsychotic to augment treatment with
clozapine. An adequate trial of such an augmentation may need to be up to 8–
10 weeks. Choose a drug that does not compound the common side effects of
clozapine. [2009]

10.11.2.1 More long-term, head-to-head RCTs of the efficacy and
safety/tolerability and patient acceptability of the available antipsychotic
drugs are required, in individuals in their first episode of schizophrenia,
testing the risk- benefit of dosage at the lower end of the recommended
dosage range. [2009]
10.11.2.2 Large-scale, observational, survey-based studies, including qualitative
components, of the experience of drug treatments for available
antipsychotics should be undertaken. Studies should include data on service
user satisfaction, side effects, preferences, provision of information and
quality of life. [2009]

10.11.2.3 Quantitative and qualitative research is required to investigate the
utility, acceptability and safety of available drugs for urgent
sedation/control of acute behavioural disturbance (including
benzodiazepines and antipsychotics), systematically manipulating dosage
and frequency of drug administration. [2009]
10.11.2.4 Further work is required on the nature and severity of antipsychotic
drug discontinuation phenomena, including the re-emergence of psychotic
symptoms, and their relationship to different antipsychotic withdrawal
strategies. [2009]
10.11.2.5 Direct comparisons between available oral antipsychotics are needed to
establish their respective risk/long-term benefit, including effects upon
relapse rates and persistent symptoms, and cost effectiveness. Trials should
pay particular attention to the long-term benefits and risks of the drugs,
including systematic assessment of side effects: metabolic effects (including
weight gain), EPS (including tardive dyskinesia), sexual dysfunction,
lethargy and quality of life. [2009]
10.11.2.6 Further RCT-based, long-term studies are needed to establish the
clinical and cost effectiveness of available depot/long-acting injectable
antipsychotic preparations to establish their relative safety, efficacy in terms
of relapse prevention, side-effect profile and impact upon quality of life.
[2009]
10.11.2.7 Further RCT-based, long-term studies are needed to establish the
clinical and cost effectiveness of augmenting antipsychotic monotherapy
with an antidepressant to treat persistent negative symptoms. [2009]
10.11.2.8 Controlled studies are required to test the efficacy and safety of
combining antipsychotics to treat schizophrenia that has proved to be poorly
responsive to adequate trials of antipsychotic monotherapy. [2009]
10.11.2.9 A randomised placebo-controlled trial should be conducted to
investigate the efficacy and post effectiveness of augmentation of clozapine
monotherapy with an appropriate second antipsychotic where a refractory
schizophrenic illness has shown only a partial response to clozapine. [2009]
investigate the efficacy and cost effectiveness of augmentation of
antipsychotic monotherapy with lithium where a schizophrenic illness has
shown only a partial response. The response in illness with and without
affective symptoms should be addressed.[2009]

investigate the efficacy and cost effectiveness of augmentation of
antipsychotic monotherapy with sodium valproate where a schizophrenic
illness has shown only a partial response. The response of illness in relation
to behavioural disturbance, specifically persistent aggression, should be
specifically addressed to determine if this is independent of effect on
potentially confounding variables, such as positive symptoms, sedation, or
akathisia. [2009]
10.11.2.12 Further controlled studies are required to test the claims that clozapine
is particularly effective in reducing hostility and violence, and the
inconsistent evidence for a reduction in suicide rates in people with
schizophrenia. [2009]

11 ECONOMIC MODEL - COST
EFFECTIVENESS OF
PHARMACOLOGICAL
INTERVENTIONS FOR PEOPLE
WITH SCHIZOPHRENIA
11.1 INTRODUCTION
This chapter has not been updated. Sections of the guideline where the evidence has
not been updated since 2009 are marked by asterisks (**2009**_**2009**).
11.1.1 Rationale for economic modelling – objectives

**2009**The systematic search of economic literature identified a number of studies
on pharmacological treatments for the management of schizophrenia which were of
varying quality and relevance to the UK setting. Results were characterised, in most
cases, by high uncertainty and various levels of inconsistency. The number of
antipsychotic medications assessed in this literature was limited and did not include
the whole range of drugs available in the UK for the treatment of people with
schizophrenia. These findings pointed to the need for de novo economic modelling
for this guideline. The objective of economic modelling was to explore the relative
cost effectiveness of antipsychotic medications for people with schizophrenia in the
current UK clinical setting, using up-to-date appropriate information on costs and
clinical outcomes, and attempting to include a wider choice of antipsychotic drugs
than that examined in the existing economic literature as well as to overcome at least
some of the limitations of previous models. Details on the guideline systematic
review of economic literature on pharmacological interventions for people with
schizophrenia are provided in Chapter 10 (Section 10.9.1).
11.1.2 Defining the economic question

The systematic review of clinical evidence covered four major areas of treating
people with schizophrenia with antipsychotic drugs: initial treatment for people
with first-episode or early schizophrenia; treatment of people with an acute
exacerbation or recurrence of schizophrenia; promoting recovery in people with
schizophrenia that is in remission (relapse prevention); and promoting recovery in
people with schizophrenia whose illness has not responded adequately to treatment
(treatment resistance). In deciding which area to examine in the economic model, the
following criteria were considered:
• quality and applicability (to the UK context) of relevant existing
economic evidence
• magnitude of resource implications expected by use of alternative
pharmacological treatments in each area

• availability of respective clinical evidence that would allow meaningful
and potentially robust conclusions to be reached that could inform
formulation of recommendations.
Based on the above criteria, the economic assessment of antipsychotic medications

aiming at promoting recovery (preventing relapse) in people with schizophrenia that
is in remission was selected as a topic of highest priority for economic analysis:
relevant existing economic evidence was overall rather poor and not directly
transferable to the UK context. Resource implications associated with this phase of
treatment were deemed major because treatment covers a long period that can
extend over a lifetime. Finally, respective clinical evidence was deemed adequate to
allow useful conclusions from economic modelling because it covered most (but not
all) of the antipsychotic medications available in the UK and was derived from a
sufficient number of trials (17) providing data on 3,535 participants.
11.2 ECONOMIC MODELLING METHODS

11.2.1 Interventions assessed
The choice of interventions assessed in the economic analysis was determined by the
availability of respective clinical data included in the guideline systematic literature
review. Only antipsychotic medications licensed in the UK and suitable for first-line
treatment aiming at preventing relapse in people with schizophrenia that is in
remission were considered. Depot/long-acting injectable antipsychotic medications
were not included in the economic analysis because they were not deemed suitable
for first- line treatment of people with schizophrenia. Consequently, the following
seven oral antipsychotic medications were examined: olanzapine, amisulpride,
zotepine, aripiprazole, paliperidone, risperidone and haloperidol. Quetiapine was
not included in the economic analysis because no respective clinical data in the area
of relapse prevention in people with schizophrenia that is in remission were
identified in the literature. In addition, haloperidol was the only FGA evaluated
because no clinical data on other FGAs were included in the guideline systematic
review. Further clinical evidence on FGAs may exist, but may have not been
identified because the guideline systematic search of the literature focused on
clinical trials of SGAs. Non-inclusion of quetiapine and other FGAs is acknowledged
as a limitation of the economic analysis.
11.2.2 Model structure

A decision-analytic Markov model was constructed using Microsoft Office Excel
2007. The model was run in yearly cycles. According to the model structure, seven
hypothetical cohorts of people with schizophrenia that is in remission were initiated
on each of the seven oral antipsychotic medications assessed (first-line
antipsychotic). The age of the population was 25 years at the start of the model, as
this is the mean age at onset of schizophrenia. Within each year, people either
remained in remission, or experienced a relapse, or stopped the antipsychotic
because of the presence of intolerable side effects, or stopped the antipsychotic for

any other reason (except relapse or presence of intolerable side effects), or died.
People who stopped the first-line antipsychotic because of the development of
intolerable side effects switched to a second-line antipsychotic. People who stopped
the first-line antipsychotic for any other reason were assumed to stop abruptly and
move to no treatment; these people remained without antipsychotic treatment until
they experienced a relapse. People discontinuing treatment because of side effects or
other reasons were assumed not to experience relapse in the remaining time of the
cycle within which discontinuation occurred. All people experiencing a relapse
stopped any antipsychotic drug that they had been receiving while in remission and
were treated for the acute episode; after achieving remission, they either returned to
their previous antipsychotic medication aiming at promoting recovery (50% of
people achieving remission), or switched to a second-line antipsychotic drug (the
remaining 50%). People initiated on a second- line antipsychotic experienced the
same events as described above. People who stopped the second-line antipsychotic
medication either because of intolerable side effects or following a relapse (50% of
people) were switched to a third-line antipsychotic drug. No further medication
switches were assumed after this point. This means that people under the third-line
antipsychotic were assumed not to stop medication because of side effects or for
other reasons, and all of them returned to this antipsychotic after treatment of
relapses. It must be noted that discontinuation of an antipsychotic because of
intolerable side effects was assumed to occur only during the first year of use of this
particular antipsychotic. Discontinuation of an antipsychotic for other reasons was
assumed to occur over each year of use, at the same rate. People under first-, second-
or third-line antipsychotic medication might experience side effects that do not lead
to discontinuation (tolerable side effects). All transitions in the model, for purposes
of estimation of costs and QALYs, were assumed to occur in the middle of each
cycle. Two different time horizons were examined (10 years and over the lifetime of
the study population), to allow exploration of the impact of long-term benefits and
risks of antipsychotic medications on their relative cost effectiveness over time. A
schematic diagram of the economic model is presented in Figure 1.
The first-line antipsychotic described in the model structure was one of the seven
oral antipsychotics evaluated in the analysis. The second-line antipsychotic
following first-line olanzapine, amisulpride, zotepine, aripiprazole, paliperidone or
risperidone was an FGA; the second-line antipsychotic following first-line
haloperidol was an SGA. The third-line antipsychotic was in all cases a depot
antipsychotic medication. In terms of costs, relapse and discontinuation and side
effect rates, the FGA used as second-line treatment was assumed to be haloperidol;
the SGA used as second-line treatment was assumed to be olanzapine; the depot
antipsychotic (third- line treatment) was assumed to be flupentixol decanoate, as this
is the most commonly used depot antipsychotic in UK clinical practice (NHS The
Information Centre, 2008c).

Figure 1: Schematic diagram of the economic model structure
Note. AP = antipsychotic.
The aim of the consideration of three lines of treatment in the model structure was
not to assess or recommend specific sequences of drugs. The model evaluated the
relative cost effectiveness between the first-line antipsychotics only. The purpose of
incorporating medication switching in the model structure was to assess the impact
of lack of effectiveness in relapse prevention (expressed by relapse rates), intolerance
(expressed by discontinuation rates because of side effects) and unacceptability
(expressed by discontinuation rates because of other reasons) of the first-line
antipsychotics on future costs and health outcomes, and to present a more realistic
sequence of events related to treatment of people with schizophrenia with
antipsychotic medication. The seven sequences of antipsychotic medications
considered in the analysis are presented in Figure 2.
11.2.3 Costs and outcomes considered in the analysis

The economic analysis adopted the perspective of the NHS and personal social
services, as recommended by NICE (NICE, 2007). Costs consisted of drug acquisition
costs, inpatient and outpatient secondary care costs, costs of primary and
community healthcare, costs of treating side effects and related future complications,
as well as costs of residential care. The measure of outcome was the QALY.

Figure 2: Sequences of antipsychotic treatment assumed in the model for each of
the seven hypothetical cohorts of people with schizophrenia followed
First-line antipsychotic Second-line antipsychotic Third-line antipsychotic

Olanzapine FGA Depot antipsychotic medication
Amisulpride FGA Depot antipsychotic medication
Zotepine FGA Depot antipsychotic medication
Aripiprazole FGA Depot antipsychotic medication
Paliperidone FGA Depot antipsychotic medication
Risperidone FGA Depot antipsychotic medication
Haloperidol SGA Depot antipsychotic medication
11.2.4 Overview of methods employed for evidence synthesis

To populate the economic model with appropriate input parameters, the available
clinical evidence from the guideline systematic review and meta-analysis needed to
be combined in a way that would allow consideration of all relevant information on
the antipsychotics assessed. The systematic review of clinical evidence in the area of
relapse prevention identified 17 trials that made pair-wise comparisons between an
SGA and another SGA, an FGA, or placebo. To take all trial information into
consideration, without ignoring part of the evidence and without introducing bias
by breaking the rules of randomisation (for example, by making ‘naive’ addition of
data across relevant treatment arms from all RCTs as described in Glenny and
colleagues (2005), mixed treatment comparison meta-analytic techniques were
employed. Mixed treatment comparison meta-analysis is a generalisation of
standard pair-wise meta-analysis for A versus B trials to data structures that include,
for example, A versus B, B versus C and A versus C trials (Lu & Ades, 2004). A basic
assumption of mixed treatment comparison methods is that direct and indirect
evidence estimate the same parameter; in other words, the relative effect between A
and B measured directly from an A versus B trial is the same with the relative effect
between A and B estimated indirectly from A versus C and B versus C trials. Mixed
treatment comparison techniques strengthen inference concerning the relative effect
of two treatments by including both direct and indirect comparisons between
treatments and, at the same time, allow simultaneous inference on all treatments
examined in the pair-wise trial comparisons while respecting randomisation
(Caldwell et al., 2005; Lu & Ades, 2004). Simultaneous inference on the relative effect
a number of treatments is possible provided that treatments participate in a single
‘network of evidence’, that is, every treatment is linked to at least one of the other
treatments under assessment through direct or indirect comparisons.
Mixed treatment comparison methods were undertaken to make simultaneous

inference for the antipsychotic drugs included in the economic analysis on the
following five parameters: probability of relapse, probability of treatment
discontinuation because of intolerable side effects, probability of treatment
discontinuation because of any other reason, probability of weight gain and
probability of acute EPS. Data on the first three parameters were analysed together

using a mixed treatment comparison ‘competing risks’ logistic regression model
appropriate for multinomial distribution of data. Data on probability of weight gain
and probability of acute EPS were analysed using two separate logistic regression
models for binomial distributions. All three models were constructed following
principles of Bayesian analysis and were conducted using Markov Chain Monte
Carlo simulation techniques implemented in WinBUGS 1.4 (Lunn et al., 2000;
Spiegelhalter et al., 2001).
11.2.5 Relapse and discontinuation data

Data on (i) relapse, (ii) drug discontinuation because of intolerable side effects and
(iii) drug discontinuation because of other reasons were taken from 17 RCTs
included in the guideline systematic review of pharmacological treatments aiming at
relapse prevention in people with schizophrenia that is in remission (details of this
review are provided in Chapter 10, Section 10.4). All 17 RCTs reported data on the
three outcomes considered in the analysis. The vast majority of the trials reported
separately on the proportions of people that discontinued treatment because of
relapse and of people discontinuing because of side effects, as well as of people
discontinuing for any other reason; overall treatment failure was defined as the sum
of these three outcomes. The outcomes were thus ‘competing’ or ‘mutually
exclusive’, in the sense that within the time frame of the trials any person who did
not remain under treatment and in remission (which would equal treatment success)
was at risk of either relapsing or stopping treatment because of side effects, or
stopping treatment because of other reasons. A small number of trials reported the
numbers of people who experienced relapse within the time frame of analysis,
without clarifying whether these people remained in the trial following relapse and
could be potentially double-counted if they discontinued treatment because of side
effects or other reasons at a later stage of the study. However, for the purpose of
analysis of clinical data and to build the economic model, data on relapse,
discontinuation because of side effects and discontinuation because of other reasons
from all 17 RCTs were treated as competing, as described above. It must be noted
that all 17 studies reported numbers of people that experienced relapse, but not the
total number of relapses per such person. It is therefore not known whether some of
the trial participants could have experienced more than one episode of relapse
during the time frame of analyses. Consequently, clinical data have been analysed
assuming that participants reported to have experienced relapse had only one
episode of relapse over the time frame of each trial. A final limitation of the data
analysis lay in the fact that the 17 RCTs used various definitions of relapse
(described in Chapter 10, Sections 10.4.4 and 10.4.5) and therefore the reported
relapse rates are not entirely comparable across studies.

Table 114: Summary of data reported in the RCTs included in the guideline systematic review on pharmacological relapse
prevention that were utilised in the economic analysis
Study Time horizon Comparators Number of people Number of people Number of people Number of people
(weeks) relapsing (m1) stopping because of stopping because in each arm (n)
side effects (m2) of other reasons
(m3)
1.BEASLEY2003 42 Placebo (1) 28 12 15 102
Olanzapine (2) 9 2 19 224
2.DELLVA1997 46 Placebo (1) 7 0 4 13

(study1) Olanzapine (2) 10 2 16 45
3.DELLVA1997 46 Placebo (1) 5 2 5 14

(study2) Olanzapine (2) 6 10 15 48
4.LOO1997 26 Placebo (1) 5 5 39 72

Amisulpride (3) 4 1 26 69
5.Cooper2000 26 Placebo (1) 21 4 24 58

Zotepine (4) 4 16 21 61
6.PIGOTT2003 26 Placebo (1) 85 13 12 155

Aripiprazole (5) 50 16 18 155
7.Arato2002 52 Placebo (1) 43 11 7 71

Ziprasidone (6) 71 19 28 206
8.KRAMER2007a 47 Placebo (1) 52 1 7 101

Paliperidone (7) 23 3 17 104
Continued

Study Time horizon Comparators Number of Number of people Number of Number of
(weeks) people relapsing stopping because people stopping people in each
(m1) of side effects (m2) because of arm (n)
other reasons
(m3)
9.SIMPSON2005 28 Olanzapine (2) 11 6 44 71
Ziprasidone (6) 8 5 33 55
10.Tran1998 52 Olanzapine (2) 87 54 170 627

(a + b + c)b Haloperidol (8) 34 20 50 180
11.STUDY-S029 52 Olanzapine (2) 28 9 26 141

Haloperidol (8) 29 14 25 134
12.Tran1997 28 Olanzapine (2) 20 17 36 172

Risperidone (9) 53 17 18 167
13.Speller1997 52 Amisulpride 5 3 2 29
(3) Haloperidol 9 5 2 31
(8)
14.Csernansky2000 52 Haloperidol (8) 65 29 80 188

Risperidone (9) 41 22 60 177
15.MARDER2003 104 Haloperidol (8) 8 0 4 30

Risperidone (9) 4 3 4 33
Note. a Participants received treatment for up to 11 months (47 weeks)

b Data from the three RCTs with study ID Tran1998a+b+c are presented together because discontinuation data were not reported
separately for each trial. The time horizon for a + b studies was 52 weeks. In study c, participants completed between 22 and 84 weeks of
therapy. For modelling purposes, the time horizon in all three studies was assumed to be 52 weeks.

Figure 3: Evidence network derived from data on relapse, treatment
discontinuation because of intolerable side effects and treatment discontinuation
for other reasons
Risperidone
Haloperidol
Amisulpride Olanzapine
Placebo
Ziprasidone
Aripiprazole
Zotepine Paliperidone
Note. Ziprasidone (in grey-shaded oval ) was considered in the mixed treatment comparison analysis because
it allowed indirect comparison between olanzapine and placebo, thus strengthening inference. However, it
was not included in the economic analysis because it is not licensed in the UK.
The time horizon of the RCTs ranged from 26 to 104 weeks. Two of the trials
assessed ziprasidone versus placebo and versus olanzapine. Ziprasidone is not
licensed in the UK and for this reason was not considered in the economic analysis;
nevertheless, data from these RCTs were utilised in the mixed treatment comparison
model because they allowed indirect comparison between olanzapine and placebo,
thus strengthening inference. Table 114 provides a summary of the data utilised in
the mixed treatment comparison competing risks model. The network of evidence
resulting from the available data is shown in Figure 3
Mixed treatment comparisons – competing risks model for relapse and

discontinuation data
A random effects model was constructed to estimate for every antipsychotic drug
evaluated the probabilities of relapse, treatment discontinuation because of
intolerable side effects and treatment discontinuation because of other reasons over
52 weeks, using data from the 17 RCTs summarised in Table 114. The data for each
trial j constituted a multinomial likelihood with four outcomes: m = 1 relapse, 2 =
discontinuation because of intolerable side effects, 3 = discontinuation because of
other reasons and 4 = none of these (treatment success). If rjm is the number observed
in each category and nj is the total number at risk in trial j, then:

𝑚 =4
𝑟𝑗,𝑚=1,2,3,4 ~𝑀𝑢𝑙𝑡𝑖𝑛𝑜𝑚𝑖𝑎𝑙 �𝑝𝑗,𝑚=1,2,3,4 , 𝑛𝑗 � 𝑤ℎ𝑒𝑟𝑒 � 𝑝𝑚 = 1
𝑚 =1
Each of the three outcomes m = 1, 2, 3 was modelled separately on the log hazard
rate scale. For outcome m, treatment k in trial j, and considering a trial j comparing
treatments k and b,
𝜃𝑗,𝑘,𝑚 = 𝜇𝑗,𝑚 + 𝛿𝑗,𝑏,𝑘,𝑚 𝐼(𝑏 ≠ 𝑘), 𝑚 = 1, 2, 3
where dj,b,k,m is the trial-specific log hazard ratio of treatment k relative to treatment
b. μj,m is the ‘baseline’ log hazard in that trial, relating to treatment b. The trial-
specific log hazard ratios were assumed to come from a normal ‘random effects’
distribution:
2
𝛿𝑗,𝑏,𝑘,𝑚 ~ 𝑁𝑜𝑟𝑚𝑎𝑙(𝑑𝑘,𝑚 − 𝑑𝑏,𝑚 , 𝜎𝑚 )
The mean of this distribution is a difference between mean relative effects dk,m and
db,m, which are the mean effects of treatments k and b respectively relative to
treatment 1, which is placebo, for outcome m. This formulation of the problem
expresses the consistency equations were assumed to hold (Lu & Ades, 2006). The
between- trials variance of the distribution was specific to each outcome m.
Vague priors were assigned to trial baselines in the estimation of relative effects
and to mean treatment effects, mj,dk,m~ N(0,1002).
A competing risks model was assumed, with constant hazards exp(θj,k,m) acting over
the period of observation Dj in years. Thus, the probability of outcome m by the end
of the observation period for treatment k in trial j was:
m =3
exp(θ j ,k ,m )[1 − exp(−∑ D j exp(θ j ,k ,m )]
p j ,k ,m ( D j ) = m =3
m =1
, m 1, 2,3
m =1
∑ exp(θ j ,k ,m )
To obtain absolute effects for use in the economic model requires an estimate of the
baseline effect in the absence of treatment. While it is desirable to allow the baseline
effects to be unconstrained so as to obtain unbiased estimates of relative effects, for
the economic model in this guideline a baseline effect that represents the trial
evidence was inputted. Therefore, a separate model was constructed for the response
to placebo, based on the eight trials with a placebo arm. The response on each
outcome was again modelled on a log hazard scale.

ξ j ,m ~ N ( B, ω m2 ), B ~ N (0,1002 )
m =3
exp(ξ j ,m )[1 − exp(−∑ D j exp(ξ j ,m ))]
p j ,m ( D j ) = m =3
m =1
, m 1, 2,3
∑ exp(ξ j ,m )
m =1
Priors for the between-trials variation were constructed as follows. First, for the
between-studies variation regarding placebo, each of the three outcomes was
assigned vague inverse Gamma priors: 1/ ω m2 ~Gamma (0.1, 0.1) . Then, it was assumed
that the variance of the treatment differences must be between zero (perfect
correlation between arms) and unity (zero correlation between arms). Thus:
σ m2 ω m2 2(1 − ρ ),
= where ρ ~ U (0,1)
For the economic analysis, the output from the model was the proportion of people
reaching each outcome by 52 weeks on treatment. The absolute log hazard
Θk,m for outcome m on treatment k was based on the mean treatment effect relative to
treatment 1 (that is, placebo) and a random sample X k,m from the distribution of
absolute log hazards on placebo:
Χ m ~N (ξ m , ω m2 )
Θ k ,m =Χ m + d k ,m
m =3
exp(Θ k ,m )[1 − exp(−∑ exp(Θ k ,m ))]
=Pk ,m = m =3
m =1
, m 1, 2,3
∑ exp(Θk ,m ) m =1
m =3
Pk ,4 = 1 − ∑ Pk ,m
m =1
Model parameters required for the economic analysis were estimated using Markov
chain Monte Carlo simulation methods implemented in WinBUGS 1.4 (Lunn et al.,
2000; Spiegelhalter et al., 2001). The first 60,000 iterations were discarded and
300,000 further iterations were run; because of high autocorrelation observed in
some model parameters, the model was thinned so that every 30th simulation was
retained. Consequently, 10,000 posterior simulations were recorded. To test whether
prior estimates had an impact on the results, two chains with different initial values
were run simultaneously. Convergence was assessed by inspection of the Gelman–
Rubin diagnostic plot.
The Winbugs code used to estimate the 52-week probabilities of (i) relapse, (ii)
treatment discontinuation because of side effects and (iii) treatment discontinuation
because of other reasons is provided in Appendix 26, followed by summary statistics

of a number of model parameters, including the log hazard ratios of all evaluated
drugs relative to placebo on the three outcomes examined and the between-trials
variation for each outcome. Results are reported as mean values with 95% credible
intervals, which are analogous to confidence intervals in frequentist statistics. Table
115 presents the mean values and 95% credible intervals of the probabilities of each
outcome for each of the drugs evaluated in the economic analysis, as well as the
probability of each treatment being the best with respect to each of the outcomes
considered. It can be seen that results for all antipsychotic drugs and all outcomes
are characterised by high uncertainty, as expressed by wide 95% credible intervals.
Goodness of fit was tested using the deviance information criterion (DIC) tool. Three
different models were tested: a fixed effects model, a random effects model
assuming the same between-trials variance of distribution for all three outcomes and
the random effects model described above, which allowed between-trials variance of
distribution specific for each outcome. The data showed a considerably worse fit in
the fixed effects model (DIC = 676.7) compared with the random effects model with
common between-trials variance for all three outcomes (DIC = 661.6) and the
random effects model with between-trials variance specific for each outcome (DIC =
659.9). Data fit well in both random effects models.
The probability of relapse and the probability of treatment discontinuation because

of other reasons over 52 weeks were assumed to apply to every (yearly) cycle of the
economic model. The probability of treatment discontinuation because of intolerable
side effects over 52 weeks was assumed to apply only to the first year following
initiation of a particular antipsychotic drug.

Table 115: Results of mixed treatment comparison analysis – competing risks
model
Treatment Probability of relapse over 52 weeks Probability that treatment is

best in reducing relapse over
Mean Lower CI Upper CI 52 weeks
Olanzapine 0.1996 0.0146 0.7222 0.078
Amisulpride 0.2988 0.0197 0.9042 0.043

Zotepine 0.1067 0.0023 0.5601 0.486
Aripiprazole 0.2742 0.0130 0.8531 0.061
Paliperidone 0.1625 0.0025 0.7008 0.270
Risperidone 0.2761 0.0182 0.8785 0.044
Haloperidol 0.3317 0.0262 0.9028 0.018
Placebo 0.4361 0.0913 0.8613 0.000
Probability of discontinuation because Probability that treatment is
of side effects over 52 weeks best in reducing discontinua-
tion because of side effects
Mean Lower CI Upper CI over 52 weeks
Olanzapine 0.0783 0.0021 0.4784 0.152
Amisulpride 0.0554 0.0006 0.3721 0.444

Zotepine 0.3821 0.0120 0.9750 0.011
Aripiprazole 0.1582 0.0026 0.7847 0.084
Paliperidone 0.3287 0.0039 0.9770 0.053
Risperidone 0.1032 0.0020 0.6735 0.134
Haloperidol 0.0922 0.0017 0.5386 0.116
Placebo 0.1094 0.0088 0.4047 0.006
Probability of discontinuation because Probability that treatment is
of other reasons over 52 weeks best in reducing discontinua-
tion because of other reasons
Mean Lower CI Upper CI over 52 weeks
Olanzapine 0.2730 0.0207 0.8596 0.030
Amisulpride 0.2435 0.0139 0.8324 0.123

Zotepine 0.2253 0.0074 0.8189 0.229
Aripiprazole 0.3520 0.0202 0.9218 0.046
Paliperidone 0.3848 0.0090 0.9479 0.105
Risperidone 0.1761 0.0086 0.7141 0.390
Haloperidol 0.2516 0.0151 0.8290 0.069
Placebo 0.2754 0.0273 0.7849 0.008
Note. Mean values and 95% credible intervals (CIs) of probabilities of (i) relapse, (ii)
treatment discontinuation because of side effects and (iii) treatment discontinuation because
of other reasons and probabilities of each treatment being the best in ranking for each of the
above outcomes (data on ziprasidone not reported – ziprasidone not considered in ranking).

Probability of relapse under no treatment
People discontinuing treatment because of other reasons and moving to no
treatment were assumed to stop treatment abruptly, and were therefore at high risk
of relapse, reaching 50%, in the first 7 months (Viguera et al., 1997). The annual
probability of relapse for no treatment (following treatment discontinuation because
of other reasons) was assumed to be equal to that estimated in the mixed treatment
comparison analysis for placebo, with the exception of the first year following
treatment discontinuation: for this year a higher probability of relapse was
estimated, taking into account the data reported in Viguera and colleagues (1997).
Probability of relapse for depot antipsychotic medication

The annual probability of relapse for the third-line depot antipsychotic medication
was taken from data reported in a Cochrane Review on flupentixol decanoate (David
et al., 1999). The reported probability (29.77%) may seem rather high; however, this
estimate was based on intention-to-treat analysis. Considering that the depot
antipsychotic was the final line of treatment in the model and no further
discontinuations (which indicate lower compliance) were allowed, the figure of
29.77% seemed reasonable and appropriate to use in the analysis, to reflect potential
non-compliance associated with depot antipsychotic medication.
11.2.6 Side effect data

The choice of side effects for consideration in the economic analysis was based on a
number of criteria, including the number of people affected in the study population,
the impact of side effects on the HRQoL, the magnitude of costs incurred by their
management and the availability of respective clinical data specific to the treatment
options assessed. Based on the above criteria, three side effects were modelled:
weight gain, acute EPS and glucose intolerance/insulin resistance as a representative
feature of the metabolic syndrome. It must be noted that acute EPS did not include
cases of tardive dyskinesia; the latter differs from acute EPS as it has lasting effects
and was not considered in the analysis. Omission of tardive dyskinesia and other
neurological side effects, as well as other side effects of antipsychotic medication that
may lead to impairments in quality of life (such as sexual dysfunction, increase in
prolactin levels, and cardiovascular and gastrointestinal side effects), is
acknowledged as a limitation of the economic analysis.
Weight gain
Data on rates of weight gain were derived from the guideline systematic review of
side effects of antipsychotic medication (details of this review are provided in
Chapter 10, Section 10.7). Only data reported as ‘number of people experiencing an
increase in weight of at least 7% from baseline’ were considered for the economic
analysis because this measure ensured a consistent and comparable definition of
weight gain across trials.
Table 114 presents a summary of the data included in the guideline systematic
review and utilised in the mixed treatment comparison analysis. Data were available

for six out of the seven antipsychotic medications evaluated in the economic analysis
(that is, olanzapine, amisulpride, aripiprazole, paliperidone, risperidone and
haloperidol). In addition, four trials that compared quetiapine with another
antipsychotic drug were considered in the mixed treatment comparison analysis:
two of the trials compared quetiapine with risperidone, one with haloperidol and
one with olanzapine. Although quetiapine was not considered in the economic
analysis because of lack of clinical data in the area of relapse prevention, quetiapine
data on weight gain were considered in the respective mixed treatment comparison
analysis as they allowed indirect comparisons across some antipsychotic
medications, thus strengthening inference. Trials comparing an SGA with an FGA
other than haloperidol were not considered in the mixed treatment comparison
analysis as data on FGAs other than haloperidol were sparse; for this reason FGAs
other than haloperidol have been treated as a class in the guideline meta-analysis.
Nevertheless, such a methodology was considered inappropriate for mixed
treatment comparison analysis. The network of evidence resulting from the available
data is shown in Figure 4.
Mixed treatment comparisons – simple random effects model for data on weight
gain
A simple random effects model was constructed to estimate the relative effect
between the k = 7 antipsychotic drugs evaluated in terms of weight gain, using data
from the 17 RCTs summarised in Table 116. The model is similar to that described by
Hasselblad (1998). The data for each trial j comprised a binomial likelihood:
𝑟𝑗𝑘 ~ Bin (𝑝𝑗𝑘 , 𝑛𝑗𝑘 )
where pjk is the probability of experiencing weight gain in trial j under treatment k, rjk
is the number of people experiencing weight gain in trial j under treatment k and njk
is the total number of people at risk in trial j under treatment k.
Treatment effects were modelled on the log-odds scale and were assumed to be
additive to the baseline treatment b in trial j:
logit(pjk) = μjb for k = b;

logit(pjk) = μjb + δjkb for k ≠ b
where μjb is the log odds of weight gain for baseline treatment b in trial j and δjkb is
the trial-specific log-odds ratio of treatment k relative to treatment b.

Table 116: Summary of data reported in the RCTs included in the guideline systematic review on weight gain (‘increase in
weight ≥7% from baseline’) that were utilised in the economic analysis
Study Time 1. Haloperidol 2. Olanzapine 3. Aripiprazole 4. Quetiapine 5. Paliperidone 6. Risperidone 7. Amisulpride

horizon (r/n) (r/n) (r/n) (r/n) (r/n) (r/n) (r/n)
(weeks)
1.LIEBERMAN2003A 24 51/132 95/131 - - - -
2.KONGSAKON2006 24 30/94 51/113 - - - -
3.StudyS029 52 23/128 46/134 - - - -
4.KANE2002 4 10/103 - 11/203 - - -
5.Arvanitis1997 6 2/52 - - 20/157 - -
6.MCQUADE2004 26 - 58/155 21/154 - - -
7.RIEDEL2007B 8 - 8/17 - 8/16 - -
8.DAVIDSON2007 6 - 25/115 - - 13/118 -
9.KANE2007A 6 - 16/123 - - 6/118 -
10.MARDER2007 6 - 23/109 - - 8/112 -
11.Conley2001 8 - 44/161 - - - 18/155
12.MARTIN2002 24 - 66/186 - - - - 39/186
13.POTKIN2003A 4 - - 22/201 - - 11/99

14.CHAN2007B 4 - - 2/49 - - 4/34
15.RIEDEL2005 12 - - - 3/22 - 1/22
16.ZHONG2006 8 - - - 35/338 - 35/334
17.Lecrubier2000 26 - - - - - 18/100 32/95

Figure 4: Evidence network for data on weight gain (defined as an increase of at
least 7% of baseline weight).
Amisulpride Paliperidone
Olanzapine
Aripiprazole Haloperidol
Risperidone Quetiapine
By taking haloperidol (treatment A) as baseline, and the true mean treatment effects
of the remaining six treatments B, C, D, etc relative to haloperidol as the basic
parameters dAB, dAC, dAD, the remaining functional parameters can be expressed in
terms of these basic parameters, for example:
dBC = dAC – dAB; dBD = dAD – dAB; etc
The trial-specific log-odds ratios for every pair of treatments XY were assumed to
come from normal random effects distributions:
δjXY ~ N (dXY, σ2)
where dXY is the true mean effect size between X and Y and σ2 the variance of the
normal distribution, which was assumed to be common in all pairs of treatments.
Vague priors were assigned to trial baselines, basic parameters and common
variance:
μjb, dAB, dAC, dAD, etc ~ N(0, 1002); σ ~ Uniform(0,2)
The results of mixed treatment comparison analysis were recorded as odds ratios
(ORs) of weight gain for each of the six antipsychotics (olanzapine, amisulpride,
aripiprazole, quetiapine, paliperidone and risperidone) versus haloperidol (which
was used as baseline). Posterior distributions were estimated using Markov chain
Monte Carlo simulation methods implemented in Winbugs 1.4 (Lunn et al., 2000;

Spiegelhalter et al., 2001). The first 60,000 iterations were discarded and 300,000
further iterations were run; because of potentially high autocorrelation, the model
was thinned so that every 30th simulation was retained. Consequently, 10,000
posterior simulations were recorded.
The Winbugs code used to estimate the ORs of weight gain for the six antipsychotic
medications versus haloperidol is presented in Appendix 26, followed by summary
statistics of a number of model parameters, including the ORs of each antipsychotic
drug considered in the mixed treatment comparison model versus haloperidol and
the between-trials variation.
Goodness of fit was tested using the residual deviance (resdev) and the deviance
information criteria (DIC) tool. The simple random effects model demonstrated a
better fit for the data (resdev = 45.06; DIC = 296.794) compared with a fixed effects
model (resdev = 63.59; DIC = 306.519).
The probability of experiencing weight gain associated with haloperidol was

calculated using data from RCTs included in the mixed treatment comparison
analysis. The studies reporting increase in weight of at least 7% following use of
haloperidol had time horizons ranging from 4 to 52 weeks. However, it was
estimated that the rate of weight gain is not constant over time and that the majority
of new cases of weight gain develop over the first 12 weeks following initiation of
any particular antipsychotic drug. For this reason, only RCTs examining haloperidol
with time horizons of up to 12 weeks were considered at the estimation of a
weighted probability of weight gain for haloperidol. Rates of experiencing at least a
7% increase in weight reported in studies of duration shorter that 12 weeks were
extrapolated to 12-week rates using exponential fit (assuming that the rate of
experiencing an increase in weight of at least 7% remained stable over 12 weeks).
The weighted average probability of weight gain for haloperidol was subsequently
calculated from these estimates. The probabilities of weight gain (px) for each of the
other antipsychotic medications included in the mixed treatment comparison
analysis were then estimated using the following formulae:
px = oddsx / (1 + oddsx)
and
oddsx = ORx,b * pb/(1 - pb)
where pb is the probability of weight gain for haloperidol, ORx,b is the odds ratio for
weight gain with each antipsychotic drug versus haloperidol as estimated in the
mixed treatment comparison analysis, and oddsx is the odds of each antipsychotic to
cause weight gain.

Table 117: Increase in weight as a side effect of antipsychotic medications: ORs
versus haloperidol, odds and absolute probabilities (mean values)
Antipsychotic OR versus Odds Probability of Source

drug haloperidol weight gain
Haloperidol 1 0.2500 0.2000 Probability based on

extrapolation of data from RCTs
with timehorizonupto12weeks
included in the guideline
systematic review
Olanzapine 2.8631 0.7158 0.4172 ORs versus haloperidol taken
Amisulpride 1.8604 0.4651 0.3175 from mixed treatment
comparison analysis (simple
Aripiprazole 0.7373 0.1843 0.1516 random effects model)
Paliperidone 1.0779 0.2695 0.2123
Risperidone 1.0895 0.2724 0.2141
provides the estimated probability of weight gain for haloperidol, the mean ORs of
each antipsychotic drug examined in economic analysis versus haloperidol as
derived from respective mixed treatment comparison analysis, as well as the
estimated odds and probability of weight gain for each antipsychotic.
The drug-specific probabilities of experiencing weight gain derived from the above
calculations were applied to the first year following initiation of a particular
antipsychotic drug. In the following years, the probability of weight gain under this
particular antipsychotic medication was assumed to be zero (for people at risk; that
is, for those who had not already experienced weight gain).
Probability of experiencing weight gain under zotepine, depot antipsychotic

medication and no treatment
The probability of experiencing weight gain for zotepine was assumed to equal the
respective probability for risperidone; the probability for the third-line depot
antipsychotic medication was assumed to equal that of haloperidol. People under no
treatment were assumed to experience no increase in their weight equalling or
exceeding 7% of their initial weight.
Acute extrapyramidal symptoms

Data on rates of acute EPS were derived from the guideline systematic review of side
effects of antipsychotic medication (details of this review are provided in Chapter 10,
Section 10.7). Of the available data, those expressing ‘need for anticholinergic
medication’ were considered for the economic analysis as this measure was thought
to capture more accurately the presence of acute EPS.
Table 118 presents a summary of the data on acute EPS included in the guideline
systematic review and utilised in the mixed treatment comparison analysis.
Table 118: Summary of data reported in the RCTs included in the guideline systematic review on acute EPS (‘need for
anticholinergic medication’) that were utilised in the economic analysis
Study Time 1. 2. 3. 4. 5. 6. 7. 8.
horizon Haloperidol Risperidone Olanzapine Zotepine Amisulpride Quetiapine Aripiprazole Paliperidone
(weeks) (r/n) (r/n) (r/n) (r/n) (r/n) (r/n) (r/n) (r/n)
1.Claus1991 12 6/22 4/22 - - - - - -

2.Mesotten1991 8 12/32 9/28 - - - - - -
3. Chouinard1993 8 15/21 29/68 - - - - - -
4.Marder1994 8 31/66 72/256 - - - - - -
5.Peuskens1995 8 67/226 201/907 - - - - - -
6.Blin1996 4 7/20 5/21 - - - - - -
7.Janicak1999 6 22/32 12/30 - - - - - -
8.Heck2000 6 10/37 11/40 - - - - - -
9.Emsley1995 6 63/84 50/99 - - - - - -
10.SCHOOLER2005 52 68/137 48/116 - - - - - -
11.Csernansky2000 52 33/188 16/177 - - - - - -
12.MARDER2003 104 26/30 23/33 - - - - - -
13.Jones1998 54 17/23 9/21 3/21 - - - - -
14.Tollefson1997 6 315/660 - 228/1336 - - - - -
15.KONGSAKON2006 24 30/94 - 24/113 - - - - -
16.LIEBERMAN2003A 24 65/125 - 21/125 - - - - -
17.Klieser1996 4 25/45 - - 6/20 - - - -
Continued

Study Time 1. 2. 3. 4. 5. 6. 7. 8.
horizon Haloperidol Risperidone Olanzapine Zotepine Amisulpride Quetiapine Aripiprazole Paliperidone
(weeks) (r/n) (r/n) (r/n) (r/n) (r/n) (r/n) (r/n) (r/n)
18.Barnas1987 7 13/15 - - 8/15 - - - -
19.Petit1996 8 62/63 - - 42/63 - - - -
20.Delcker1990 6 13/20 - - - 11/21 - - -
21.Moller1997 6 54/96 - - - 28/95 - - -
22.Puech1998 4 26/64 - - - 45/194 - - -
23.Speller1997 52 25/31 - - - 10/29 - - -
24.Emsley1999 8 17/145 - - - - 3/143 - -
25.KANE2002 4 30/103 - - - - - 23/203 -
26.KASPER2003 52 245/430 - - - - - 196/853 -
27.Conley2001 8 - 61/188 53/189 - - - - -
28.Tran1997 28 - 55/167 34/172 - - - - -
29.Fleurot1997 8 - 26/113 - - 35/115 - - -
30.Lecrubier2000 26 - 47/158 - - 36/152 - - -
31.ZHONG2006 8 - 23/334 - - - 19/338 - -
32.RIEDEL2005 12 - 9/22 - - - 2/22 - -
33.CHAN2007B 4 - 14/34 - - - - 12/49 -
34.SIROTA2006 26 - - 6/21 - - 5/19 - -
35.KANE2007A 6 - - 10/128 - - - - 14/123
36.MARDER2007 6 - - 13/109 - - - - 10/112

Data on all seven antipsychotic medications evaluated in the economic analysis
(olanzapine, amisulpride, zotepine, aripiprazole, paliperidone, risperidone and
haloperidol) were available. In addition, four trials that compared quetiapine with
another antipsychotic drug were considered in the mixed treatment comparison
analysis: two of the trials compared quetiapine with risperidone, one with
haloperidol and one with olanzapine. Although quetiapine was not considered in
the economic analysis owing to lack of clinical data in the area of relapse prevention,
quetiapine data on acute EPS were considered in the respective mixed treatment
comparison analysis as they allowed indirect comparisons across drugs, thus
strengthening inference. Trials comparing an SGA with an FGA other than
haloperidol were not considered in the mixed treatment comparison analysis as data
on FGAs other than haloperidol were sparse; for this reason FGAs other than
haloperidol have been treated as a class in the guideline meta-analysis. Nevertheless,
such a methodology was considered inappropriate for mixed treatment comparison
analysis. The network of evidence constructed based on the available data is
demonstrated in Figure 5.
Mixed treatment comparisons full random effects model for acute extrapyramidal
side-effects data
A full random effects model was constructed to estimate the relative effect between
the k = 8 antipsychotics evaluated in terms of development of acute EPS, using data
from the 36 RCTs summarised in Table 118. The model is similar to that described
above, utilised for the mixed treatment comparison analysis of data on weight gain,
but takes into account the correlation structure induced by a three-arm trial (Jones,
1998) included in the 36 RCTs; this model structure relies on the realisation of
Figure 5: Evidence network for data on acute EPS (expressed as need for
anticholinergic medication)
Risperidone
Aripiprazol Quetiapine
Amisulprid
Zotepine
Paliperidon
Olanzapine Haloperidol
Note. Quetiapine (in grey-shaded oval) was considered in the mixed treatment comparison
analysis because it allowed indirect comparisons between a number of medications, thus
strengthening inference. However, it was not included in the economic analysis because no
clinical data in the area of relapse prevention for people with schizophrenia that is in remission
were available for quetiapine.

the bivariate normal distribution as a univariate marginal distribution and a
univariate conditional distribution (Higgins & Whitehead, 1996):
𝑥₁ 𝜇₁ 𝜎² 𝜎²/2
If �𝑥₂� ~ N ��𝜇₂� , � ��
𝜎²/2 𝜎²
1 3
then x₁ ~ N (μ₁ , σ²), and 𝑥₂⃓ 𝑥₁ ~ N (μ₂ + (𝑥₁ - μ₁), σ²)
2 4
The results of this mixed treatment comparison analysis were also recorded as ORs
of developing acute EPS for each of the seven antipsychotic drugs (olanzapine,
amisulpride, aripiprazole, zotepine, quetiapine, paliperidone and risperidone)
versus haloperidol (which was again used as baseline). Posterior distributions were
estimated using Markov chain Monte Carlo simulation methods implemented in
Winbugs 1.4 (Lunn et al., 2000; Spiegelhalter et al., 2001). The first 60,000 iterations
were discarded, and 300,000 further iterations were run; because of potentially high
autocorrelation, the model was thinned so that every 30th simulation was retained.
Consequently, 10,000 posterior simulations were recorded.
The Winbugs code used to estimate the ORs of developing acute EPS for the seven
antipsychotic medications versus haloperidol is presented in Appendix 26, followed
by summary statistics of a number of model parameters, including the OR of each
antipsychotic drug considered in the mixed treatment comparison model versus
haloperidol and the between-trials variation. The resdev of the model was 75.93.
The probability of experiencing acute EPS for haloperidol was calculated using data
from RCTs included in the mixed treatment comparison analysis. The studies
reporting the need for anticholinergic medication following use of haloperidol had
time horizons ranging from 4 to 104 weeks. However, it was estimated that the rate
of developing acute EPS is not constant over time and that the majority of new cases
of acute EPS develop over the first 8 weeks following initiation of any particular
antipsychotic drug. For this reason, only RCTs examining haloperidol with time
horizons of up to 8 weeks were considered at the estimation of a weighted
probability of acute EPS for haloperidol. Rates of acute EPS reported in studies of
duration shorter that 8 weeks were extrapolated to 8-week rates using exponential fit
(assuming that the rate of development of acute EPS remained stable over 8 weeks).
The weighted average probability of acute EPS for haloperidol was subsequently
calculated from these estimates. The probability of acute EPS (px) for each of the
other antipsychotic medications included in the mixed treatment comparison
analysis was then estimated using the following formulae:
px = oddsx / (1 + oddsx)
and
oddsx = ORx,b * pb/(1 - pb)

where pb is the probability of acute EPS for haloperidol, ORx,b the odds ratio for acute
EPS of each antipsychotic medication versus haloperidol as estimated in the mixed
treatment comparison analysis, and oddsx the odds of each antipsychotic leading to
development of acute EPS.
Table 119 provides the estimated probability of weight gain for haloperidol, the
mean ORs of each antipsychotic drug examined in economic analysis versus
haloperidol as derived from respective mixed treatment comparison analysis, as well
as the estimated odds and probability of weight gain for each antipsychotic.
The drug-specific probabilities of developing acute EPS derived from the above
calculations were applied to the first year following initiation of a particular
antipsychotic drug. In the following years, the probability of developing acute EPS
under this particular antipsychotic medication was estimated to be 10% of the
probability applied to the first year.
Probability of developing acute extrapyramidal side effects under depot

antipsychotic medication and no treatment
The probability of developing acute EPS under the third-line depot antipsychotic
medication was taken from data reported in a Cochrane Review on flupentixol
decanoate (David et al., 1999). People under no treatment were assumed to develop
no acute EPS.
Glucose intolerance/insulin resistance and diabetes

Glucose intolerance/insulin resistance was modelled as a representative feature of
the metabolic syndrome, the incidence of which is high in people taking
antipsychotic
Table 119: Development of acute EPS as a side effect of antipsychotic medications:

ORs versus haloperidol, odds and absolute probabilities (mean values)
Antipsychotic OR versus Odds Probability of Source

drug haloperidol weight gain
Haloperidol 1 1.1586 0.5367 Probability based on extrapolation

of data from RCTs with time
horizon up to 8weeks included in
the guideline systematic review
Olanzapine 0.2631 0.3048 0.2336 ORs versus haloperidol taken from

Amisulpride 0.3993 0.4626 0.3163 mixed treatment comparison
analysis (full random effects model)
Zotepine 0.1476 0.1710 0.1461
Aripiprazole 0.2517 0.2916 0.2258
Paliperidone 0.2983 0.3456 0.2569
Risperidone 0.4743 0.5495 0.3546

medication. The metabolic syndrome is a predictor of type-2 diabetes and coronary
heart disease. Both conditions are associated with a number of events and
complications that cause significant impairment in the HRQoL and incur substantial
healthcare costs. Because there is a high correlation between the two conditions, it
was decided to only model events (complications) resulting from the development of
diabetes mellitus to avoid the double-counting of health events and the
overestimation of the (negative) impact of metabolic syndrome on the cost
effectiveness of antipsychotic drugs. Modelling health events as complications of
diabetes was preferred to linking them to coronary heart disease because estimates
of the incidence of diabetes complications have been reported in the literature,
having been derived from a large prospective cohort study of people with diabetes
mellitus in the UK (Stratton et al., 2000).
The relationship between specific antipsychotic medications, risk for metabolic

syndrome and the development of type-2 diabetes has not been fully explored and
relevant data that are appropriate for modelling are sparse. A systematic review of
the metabolic effects of antipsychotic medications concluded that antipsychotics
associated with greatest increases in body weight were also associated with a
consistent pattern of clinically significant insulin resistance (Newcomer & Haupt,
2006). The authors noted that correlations between change in weight and change in
plasma glucose values were weaker overall than correlations between weight change
and change in insulin resistance, and that unchanged plasma glucose levels did not
preclude clinically significant increases in insulin resistance. The results of the
review indicated that the relative risk for diabetes mellitus during antipsychotic
medication use generally matched the rank order of weight-gain potential for the
different antipsychotics, although a significant minority of people taking
antipsychotics might experience glucose dysregulation independent of weight gain.
A systematic review and meta-analysis of studies comparing the risk for diabetes
between SGAs and FGAs in people with schizophrenia and related psychotic
disorders found that SGAs led to a greater risk for diabetes compared with FGAs
(Smith et al., 2008). Besides being associated with impaired glucose levels and
insulin resistance, antipsychotic drugs have been shown to lead directly to
development of diabetes shortly after their initiation by people with schizophrenia
(Saddichha et al., 2008; van Winkel et al., 2006; van Winkel et al., 2008).
Given that available data on the risk for glucose intolerance and/or diabetes
associated with specific antipsychotic drugs are limited, the probability of
developing glucose intolerance/insulin resistance (associated with greater future
risk for developing diabetes) and the probability of developing diabetes directly in
the first year of antipsychotic use were estimated as follows: first, estimates on these
two probabilities specific to haloperidol were made, based on reported data in
published literature. Second, drug-specific probabilities of weight gain, estimated as
described in the previous section, were used to calculate relative risks of weight gain
for each SGA included in the analysis versus haloperidol. Relative risks for weight
gain were assumed to be equal to relative risks for developing glucose
intolerance/insulin resistance and diabetes because existing evidence suggested a

high correlation between increase in weight and insulin resistance, as discussed
above (Newcomer & Haupt, 2006). Finally, relative risks of each SGA versus
haloperidol were multiplied by the haloperidol-specific estimated probabilities of
developing glucose intolerance/insulin resistance and diabetes to obtain respective
probabilities for each SGA assessed in the economic analysis. The resulting
estimates, based on the correlation between glucose intolerance/risk for diabetes
and weight gain, may be potentially conservative because an additional mechanism
leading to glucose dysregulation, independent of weight increases, appears to exist
(Newcomer & Haupt, 2006). On the other hand, the fact that the rank order of
relative risk for diabetes has been shown to match the rank order of weight-gain
potential for the different antipsychotics, according to findings of the same study,
does not guarantee that the relative risk of developing intolerance/insulin resistance
and diabetes of each SGA versus haloperidol is actually equal to their in-between
relative risk of weight-gain. The described method for estimating absolute
probabilities for developing intolerance/insulin resistance and diabetes for each
SGA in the model was deemed necessary because of a lack of other appropriate data,
but is acknowledged as a limitation of the economic analysis.
The estimated probability of directly developing diabetes during the first year of
initiation of haloperidol was based on respective rates reported in the literature for
people with schizophrenia under antipsychotic medication (van Winkel et al., 2008).
Since these studies examined populations initiated on a number of antipsychotics,
including SGAs, and the risk for developing diabetes is known to be higher for SGAs
compared with FGAs (Smith et al., 2008), the probability of developing diabetes
within the first year of initiation of haloperidol was estimated to be lower than the
respective figures reported in the literature associated with use of antipsychotics
generally. Similarly, the probability of glucose intolerance/insulin resistance within
the first year of initiation of haloperidol was estimated taking into account relevant
data identified in the guideline systematic review of clinical evidence. The resulting
estimates for haloperidol that were used in the economic analysis were 2% (first year
probability of developing diabetes) and 15% (first year probability of developing
glucose intolerance/insulin resistance).
The resulting probabilities of developing diabetes/glucose intolerance for all

antipsychotics following the methodology described above, and the ranking of
antipsychotics in terms of risk for diabetes, were consistent with evidence suggesting
that olanzapine is strongly associated with diabetic events while aripiprazole,
risperidone and haloperidol are poorly associated with such events (Dumouchel et
al., 2008).
The probability of developing diabetes directly was applied only to the first year of
initiation of any particular antipsychotic. Similarly, it was assumed that
development of glucose intolerance/insulin resistance occurred only within the first
year of initiation of any specific drug. People who did not develop insulin resistance
within the first year of initiation of a particular antipsychotic were assumed to
develop no insulin resistance in the following years, provided that they remained on

the same drug. However, insulin resistance that developed within the first year of
initiation of a specific antipsychotic was assumed to be permanent and to result in an
increased risk for diabetes over a lifetime. The annual transition probability from
impaired glucose tolerance to developing diabetes was taken from Gillies and
colleagues (2008). It is acknowledged that applying the probabilities of developing
diabetes and insulin resistance only to the first year of initiation of any particular
antipsychotic is likely to be conservative and to underestimate the impact of the
metabolic syndrome on the relative cost effectiveness of antipsychotics. On the other
hand, insulin resistance that developed within the first year of initiation of a
particular antipsychotic was assumed to be permanent and to lead to a lifetime risk
of developing diabetes.
Complications from diabetes

The probabilities of complications following development of diabetes were
estimated based on data reported in the UKPDS (Stratton et al., 2000). This was a 20-
year prospective study that recruited 5,102 people with type-2 diabetes in 23 clinical
centres based in England, Northern Ireland and Scotland. The study reported
incidence rates of complications for different levels of haemoglobin A1C
concentration (Hgb A1C). Annual probabilities of complications were estimated
based on the available data, assuming that 20% of people in the model had Hgb A1C
7 to <8%, 30% of people had 8 to <9%, 30% of people had 9 to <10% and 20% of
people had ≥10%. These assumptions took account of the clinical experience of the
GDG, according to whom, people with schizophrenia in general do not have good
glycaemic control. Incidence of complications in Stratton and colleagues (2000) were
provided as aggregate figures of fatal and non-fatal events for each complication. To
estimate the probability of fatal and non-fatal events for each complication
separately in the economic model, the reported overall incidence of deaths related to
diabetes at each level of Hgb A1C was applied to the reported incidence of each
complication at the same Hgb A1C level to estimate the proportion of fatal events
reported for each complication.
11.2.7 Mortality estimates

The risk of death is higher in people with schizophrenia than in the general
population (McGrath et al., 2008). Transition to death in the model occurred as a
result of suicide or other reasons, including increased physical morbidity
characterising people with schizophrenia that leads to increased mortality. It was
assumed that the risk of death was independent of specific antipsychotic drug use,
owing to lack of sufficient data to support the opposite hypothesis. Instead, all
people in the model were subject to increased mortality relative to the general
population, common to all antipsychotic drugs. To calculate the number of deaths
occurring each year, the increased standardised mortality ratio (SMR) observed in
people with schizophrenia (McGrath et al., 2008) was multiplied by the age- and
gender-specific mortality rates for people aged 25 years and above in the general
population in England and Wales (Office for National Statistics, 2008). The number
of deaths was calculated on the basis that the study population (people with
schizophrenia) had a male to female ratio of 1.4 to 1 (McGrath, 2006).

Death was assumed to occur in the middle of every year (cycle); this means that over
the year death occurred, people incurred half of the costs and gained half of the
QALYs they were expected to incur and gain, respectively, had they not died.
11.2.8 Utility data and estimation of quality-adjusted life years

To express outcomes in the form of QALYs, the health states of the economic model
needed to be linked to appropriate utility scores. Utility scores represent the HRQoL
associated with specific health states on a scale from 0 (death) to 1 (perfect health);
they are estimated using preference-based measures that capture people’s
preferences on, and perceptions of, HRQoL in the health states under consideration.
Systematic review of published utility scores for people with

schizophrenia
The systematic search of the literature identified six studies that reported utility
scores for specific health states and events associated with schizophrenia (Chouinard
& Albright, 1997; Cummins et al., 1998; Glennie, 1997; Lenert et al., 2004; Revicki et
al., 1996; Sevy et al., 2001).
Chouinard and Albright (1997) generated health states using data on PANSS scores
from 135 people with schizophrenia participating in a Canadian multicentre RCT of
risperidone versus haloperidol. Cluster analysis identified three clusters that
included 130 of the participants with mild, moderate and severe symptomatology. A
health-state profile was described for each cluster, including additional information
on adverse events, obtained by assessing the average scores of Extrapyramidal
Symptom Rating Scale (ESRS) subscales of parkinsonism, dyskinesia and dystonia in
each treatment group. Subsequently, 100 psychiatric nurses in the US were asked to
assign utility values to each of the three health states using standard gamble (SG)
methods.
Glennie (1997) described the development of health-state profiles specific to

antipsychotic medications, according to average PANSS scores reported in
risperidone trials included in a systematic review. The impairment in HRQoL caused
by the need for hospitalisation and the presence of EPS were also considered. In this
case, seven people with schizophrenia in Canada who were in a stable state were
asked to value the generated health states using the SG technique.
Lenert and colleagues (2004) valued health states associated with schizophrenia
constructed from the results of principal component analysis of PANSS scores; the
scores were obtained from people with schizophrenia participating in a large multi-
centre effectiveness trial conducted in the US. This analysis led to the clustering of
types of symptoms and the final development of eight health states describing
different types and severity of schizophrenia symptoms. Moreover, the presence of
common adverse events from antipsychotic medication was taken into account at
valuation. The resulting health states were valued by a sample of 441 people from
the general US population using the SG technique.

Revicki and colleagues (1996) developed five hypothetical health states (vignettes)
describing various levels of schizophrenia symptoms, functioning and well-being in
inpatient and outpatient settings, based on relevant descriptions available in the
medical literature and expert opinion. The health states were subsequently valued
by three different groups of people in the UK, using different valuation techniques:
49 people with schizophrenia in remission and their carers rated the health states
using categorical rating scales (RS) and paired comparisons (PC); a number of
psychiatrists valued the health states using categorical RS and SG techniques. The
study reported the psychiatrist-derived utility scores using SG, as well as the utility
scores derived from people with schizophrenia and their carers using PC.
Cummins and colleagues (1998) linked health states observed in people with
schizophrenia participating in an international RCT of olanzapine versus haloperidol
with specific health states generated using the IHRQoL. The methodology used to
link these two different sets of health state profiles was not clearly described.
IHRQoL is a generic measure of HRQoL, consisting of three dimensions: disability,
physical distress and emotional distress (Rosser et al., 1992). The composite health
states derived from this generic measure have been valued using the SG method.
However, detailed description of the methods of valuation has not been made avail-
able and no other application of this instrument has been identified in the literature
(Brazier, 2007b).
Finally, Sevy and colleagues (2001) reported valuations of people with schizophrenia
for a large number of side effects resulting from antipsychotic medication, using SG
methods. The purpose of the study was to assess the relationship between the utility
values obtained and the study population’s willingness to pay to remove such side
effects. The resulting scores were reported unadjusted because death was not used
as anchor value ‘zero’ and are therefore not appropriate for use in economic
modelling.
Table 120 summarises the methods used to derive health states and subsequent
utility scores associated with schizophrenia health states and events, as well as the
results of the first five studies described above, because these reported utility scores
that could potentially be used in the guideline’s economic analysis.
In addition to the above studies, a number of studies reported utility scores for
people with schizophrenia that were generated using generic preference-based
measures of HRQoL (Kasckow et al., 2001; Knapp et al., 2008; König et al., 2007;
Lewis et al., 2006b; Sciolla et al., 2003; Strakowski et al., 2005; Tunis et al., 1999).
However, any utility scores reported in these studies expressed the overall HRQoL
of the study population and were not linked to specific health states; consequently,
they were not useful for economic modelling.
König and colleagues (2007) assessed and valued the HRQoL of people with
schizophrenic, schizotypal or delusional disorders using the EQ-5D. They concluded

that EQ-5D had reasonable validity in this group of people, but its association with
the positive subscale of PANSS was rather weak. For this reason it was suggested
that EQ-5D be used in combination with disease-specific instruments in such
populations so that all aspects of HRQoL be captured. The study did not report
utility scores relating to specific health states experienced by the study population.
Lewis and colleagues (2006b) evaluated the cost effectiveness of FGAs versus SGAs,
and clozapine versus SGAs, in people with schizophrenia responding poorly to, or
being intolerant of, current antipsychotic treatment in two RCTs conducted in the
UK (CUtLASS Bands 1 and 2). Health benefits from treatment were determined by
measuring the participants’ HRQoL using the EQ-5D at various points in the trials.
Knapp and colleagues (2008) also obtained EQ-5D scores from outpatients with
schizophrenia participating in a European multicentre observational study to
evaluate the cost effectiveness of olanzapine versus other oral and depot
antipsychotics. In both of the above economic studies, the obtained EQ-5D scores
were not attached to specific health states and therefore could not be applied to the
health states described in the guideline economic analysis.
Sciolla and colleagues (2003) assessed the HRQoL of outpatients with schizophrenia
aged over 45 years using the 36-item Short-Form health survey (SF-36). The authors
stated that SF-36 adequately measured the impairment in HRQoL associated with
schizophrenia in middle aged and older people. Strakowski and colleagues (2005)
and Tunis and colleagues (1999) reported SF-36 scores in people with schizophrenia
who participated in two different clinical trials of olanzapine versus haloperidol;
both studies reported SF-36 scores at baseline and at end of treatment for each
treatment group. None of the three studies that used the SF-36 linked the obtained
scores to specific health states associated with schizophrenia; thus the data reported
were not useful in the guideline economic analysis.
Kasckow and colleagues (2001) measured the quality of life of inpatients and
outpatients with schizophrenia using the Quality of Well-Being Scale (QWB).
Although hospitalisation and high levels of positive symptoms were shown to be
associated with lower QWB scores, no health states that could be used in the guide-
line economic analysis were specified and linked with QWB-generated utility scores.

Table 120: Summary of studies reporting utility scores relating to specific health states and events associated with
schizophrenia
Study Definition of health states Valuation method Population valuing Results
Chouinard Based on cluster analysis of PANSS scores SG 100 psychiatric Mild health state: 0.61
& Albright, combined within formation from data on nurses in the US Moderate healthstate: 0.36
1997 ESRS subscales of parkinsonism, dyskinesia Severe healthstate: 0.29
and dystonia, all obtained from 135 people
with schizophrenia in Canada who
participated in a multicentre three-arm RCT
comparing risperidone versus haloperidol
versus placebo
Cummins Health states of people with schizophrenia SG Unclear Response – no EPS: 0.960
et al., 1998 participating in a RCT linked with health Response – EPS: 0.808
states generated using the IHRQoL Need for acute treatment/relapse –
No EPS: 0.762
Need for acute treatment/relapse –
EPS: 0.631
Glennie, Based on average scores from each of the SG 7people with stable Mild delusional symptoms –
1997 three PANSS subscales (positive, negative and schizophrenia in Risperidone: 0.89
general psychopathology) reported in Canada Mild delusional symptoms –
risperidone trials included in a systematic haloperidol: 0.86
review; need for hospitalisation and presence Moderate delusional symptoms: 0.82
of EPS also considered Hospitalisation: −0.07
Presence of EPS: −0.07
Lenert et Based on principal component analysis SG 441 people from US Mild (all areas low): 0.88
al., 2004 followed by cluster analysis of PANSS scores general population Moderate type I (negative
(positive, negative and general predominant): 0.75
psychopathology subscales) obtained from Moderate type II (positive
people with schizophrenia participating in predominant): 0.74
Continued

Study Definition of health states Valuation method Population valuing Results
An effectiveness trial in the US; presence Severe type I (negative predominant): 0.63
of adverse events from medication Severe type II (positive and cognitive
also considered predominant): 0.65
Severe type III (negative and cognitive
predominant): 0.53
Severe type IV (positive predominant):
0.62
Extremely severe (all symptoms high): 0.42
Orthostatic hypotension: −0.912%
Weight gain: −0.959%
Tardive dyskinesia: −0.857%
Pseudo-parkinsonism: −0.888%
Akathisia: −0.898%
Revicki et Vignettes based on medical literature SG UK psychiatrists Outpatient, excellent functioning: 0.83
al., and expert opinion Outpatient, good functioning: 0.73
1996 Outpatient, moderate functioning: 0.70
Outpatient, negative symptoms: 0.60
Inpatient, acute positive symptoms: 0.56
PC 49 people with Outpatient, excellent functioning: 0.77

schizophrenia in Outpatient, good functioning: 0.57
remission in the UK Outpatient, moderate functioning: 0.49
PC Carers of people Outpatient, excellent functioning: 0.69

with schizophrenia Outpatient, good functioning: 0.51
in the UK Outpatient, moderate functioning: 0.44

NICE recommends the EQ-5D as the preferred measure of HRQoL in adults for use
in cost-utility analysis. NICE also suggests that the measurement of changes in
HRQoL should be reported directly from people with the condition examined, and
the valuation of health states should be based on public preferences elicited using a
choice-based method, such as time trade-off (TTO) or SG, in a representative sample
of the UK population. At the same time, it is recognised that EQ-5D data may not be
available or may be inappropriate for the condition or effects of treatment (NICE,
2008a).
None of the studies summarised in Table 120 derived utility values using EQ-5D
scores valued from members of the UK general population. Three of the five studies
generated health states based on analysis of condition-specific PANSS scores
(Chouinard & Albright, 1997; Glennie, 1997; Lenert et al., 2004). Valuations in these
three studies were made by healthcare professionals in the US (Chouinard &
Albright, 1997), by people with schizophrenia in Canada (Glennie, 1997) or by
members of the public in the US (Lenert et al., 2004). All three studies used the SG
technique. Revicki and colleagues (1996) developed health states based on vignettes,
valued by people with schizophrenia and their carers using RS or PC, or by
psychiatrists using SG. Finally, Cummins and colleagues (1998) linked health states
associated with schizophrenia with health states generated using the IHRQoL.
Although the last study used a generic measure to describe health states associated
with schizophrenia, the methodology adopted in developing and valuing health
states was not clear.
A comparison of data from the three studies that analysed PANSS scores to generate
utility scores illustrated that Glennie (1997) reported the most conservative
difference in utility scores between health states (difference between moderate and
mild states 0.04–0.07; no severe state valued); Chouinard and Albright (1997)
reported the greatest differences in utility between health states (difference between
moderate and mild states 0.25; between severe and mild states 0.32); and Lenert and
colleagues (2004) reported moderate changes in utility between health states
(difference between moderate and mild states 0.13–0.14; between severe and mild
states 0.22–0.35; and between very severe and mild states 0.46). It was therefore
decided to use utility data from Lenert and colleagues (2004) in the base-case
analysis and data from the other two studies that utilised PANSS scores (Chouinard
& Albright, 1997; Glennie, 1997) in sensitivity analysis. The data by Lenert and
colleagues (2004) were selected for the base-case analysis for a number of reasons:
they were comprehensive, covering a wide range of health states of varying types
and severity of symptoms; the described health states were derived from principal
component analysis of condition-specific PANSS scores; the methodology was
described in detail; the valuations were made by members of the general population
using SG (although the population was from the US and not the UK); detailed utility
data for a number of adverse events associated with antipsychotic medication were
also reported; the study provided comprehensive data for linking PANSS scores to
specific health states and subsequently to utility scores so that, apart from modelling

exercises, these data may be used in cost-utility analyses conducted alongside
clinical trials measuring PANSS scores, thus increasing comparability across
economic evaluations of antipsychotic treatments for people with schizophrenia.
There is at least one example where these data have been used in a cost-utility
analysis undertaken alongside effectiveness trials (Rosenheck et al., 2006).
Development of health states from condition-specific instruments, such as PANSS,
may be appropriate for people with schizophrenia because these are likely to capture
more aspects of the HRQoL relating to emotional and mental status; they may also
be more sensitive for a given dimension (Brazier, 2007a). Generic measures, such as
EQ-5D, could miss some dimensions of HRQoL associated with mental symptoms.
EQ-5D has been demonstrated to associate weakly with the positive subscale of
PANSS. For this reason, it has been suggested that EQ-5D be used in combination
with disease-specific instruments in people with schizophrenia (König et al., 2007).
The data reported in Revicki and colleagues (1996) were not considered further
because they were based on vignettes, were not valued by members of the public
and, in two of the participating groups, valuations were not made using choice-
based methods. Data from Cummins and colleagues (1998) were also excluded from
further consideration because the methods used for their derivation were not clearly
reported.
Linking utility scores to health states of remission and relapse

To link the model states of remission and relapse with the utility scores reported for
PANSS-generated health states in Lenert and colleagues (2004), the GDG estimated
that the HRQoL of people in remission (model state) corresponded by 40% to
HRQoL in the (PANSS-generated) mild state and by 60% to HRQoL in the moderate
state (30% in moderate state type I and 30% in moderate state type II); the HRQoL of
people in relapse corresponded by 60% to HRQoL in the severe state type IV and by
40% to HRQoL in the very severe state.
The GDG estimated that the decrement in HRQoL of people in schizophrenia while
in acute episode (relapse) lasted for 6 months.
Utility scores for acute extrapyramidal symptoms and weight gain

The utility scores for acute EPS and weight gain were also taken from Lenert and
colleagues (2004). The reduction in HRQoL caused by acute EPS corresponded to
that reported for pseudo-parkinsonism and was estimated to last for 3 months, after
which significant improvement in acute EPS symptoms was estimated to occur
(either spontaneously after dose adjustment or following treatment). The reduction
in HRQoL caused by weight gain was permanent because an increase in weight
following use of antipsychotic medication was estimated to remain over a lifetime.
Utility scores for diabetes complications

Disutility owing to complications from diabetes was taken from the UKPDS (Clarke
et al., 2002). Utility scores in this study were generated using patient-reported EQ-

5D scores; these were subsequently valued using EQ-5D UK tariff values. Disutility
of diabetes without complications was not considered in the economic model as it
was estimated to be negligible when compared with the impairment in HRQoL
caused by schizophrenia.
11.2.9 Cost data

Costs associated with pharmacological treatment of people with schizophrenia and
related events were calculated by combining resource-use estimates with respective
national unit costs. Costs of the relapse and remission states consisted of relevant
drug acquisition costs, outpatient, primary and community care costs, costs of
treating acute episodes (relapse state only) and residential care costs. People under
no treatment (following treatment discontinuation for reasons other than relapse or
presence of intolerable side effects) were assumed to incur no costs until they
experienced a relapse. Costs associated with baseline measurements and laboratory
tests for monitoring purposes were omitted from the analysis, because they were
estimated to be the same for all antipsychotic medications evaluated. All costs were
uplifted to 2007 prices using the Hospital and Community Health Services (HCHS)
Pay and Prices Index (Curtis, 2007). Costs were discounted at an annual rate of 3.5%
annually, as recommended by NICE (NICE, 2008a).
Drug acquisition costs

Drug acquisition costs were taken from BNF 56 (British Medical Association and the
Royal Pharmaceutical Society of Great Britain, 2008), with the exception of the cost of
risperidone which was taken from the Electronic Drug Tariff (NHS Business Services
Authority & Prescription Pricing Division, 2008) because risperidone recently
became available in generic form but BNF 56 has not captured this information. The
daily dosage of antipsychotic drugs was based on the national average daily
quantity (ADQ) values reported by the NHS (NHS The Information Centre, 2008a).
In cases where no ADQ values were available, the average daily quantity was
estimated based on BNF guidance. Some of the reported doses were slightly
adjusted to match tablet/injection doses and usual injection intervals. The ADQs and
the drug acquisition cost, as well as the monthly ingredient cost for each drug
included in the analysis, are reported in Table 121. Annual drug acquisition costs for
people experiencing relapse were different because use of antipsychotic medication
for relapse prevention was assumed to be interrupted during the acute episode and
replaced with another antipsychotic (olanzapine) over this period of relapse.
Outpatient, primary and community care costs

Estimates on resource use associated with outpatient, primary and community care
were based on data reported in a UK study (Almond et al., 2004). The study collected
information on healthcare resource use from 145 people with schizophrenia
randomly selected from psychiatric caseloads drawn from urban and suburban areas
of Leicester. Of the sample, 77 had experienced a recent relapse, defined as re-
emergence or aggravation of psychotic symptoms for at least 7 days during the 6
months prior to the study (‘relapse group’); the remaining 68 had not experienced
such a relapse in the 6 months before the initiation of the study (‘non-relapse
group’). Healthcare resource use for each group over 6 months was collected
prospectively from case notes and interviews with the study participants. The study
also reported
Table 121: ADQs, drug acquisition costs and estimated monthly ingredient costs
of antipsychotic medications included in the economic model
Drug ADQ Unit Unit cost (BNF 56, September 2008) Monthly cost
Amisulpride 400 mg Generic 400 mg, 60-tab = £114.45 £57.23
Haloperidol 8 mg Generic 1.5 mg, 28-tab = £2.84; 5 mg, £14.35
28 = £7.71; 10 mg, 28 = £9.06
Olanzapine 10 mg Zyprexa 10 mg, 28-tab = £79.45; £85.13
15 mg, 28-tab = £119.18
Aripiprazole 15 mga Abilify 15 mg, 28-tab = £101.63 £108.89
Paliperidone 9 mga Invega 9 mg, 28-tab = £145.92 £156.34
Risperidone 5 mg Generic 1 mg, 60-tab = £28.38; £67.52
4 mg, 60-tab = £106.65b
Zotepine 200 mg Zoleptil 100 mg, 90-tab = £94.55 £63.03
Flupentixol 3.6 mg Depixol Conc. 100 mg/mL, 1-mL £6.70
decanoate amp = £6.25 (administered every
4 weeks)
Note. a ADQ data available–daily dosage estimated based on BNF guidance.
b Based on the Electronic Drug Tariff as of 1 December 2008 (NHS, Business Services Authority, 2008).
inpatient care resource use for the two groups, but these data were not utilised in the
economic model. It is acknowledged that the data reported in this study are not very
recent (the study was conducted in the 1990s), but no more up-to-date data that were
appropriate to inform the economic analysis were identified in the literature.
It was assumed that, over 1 year, people in the remission state in the model
(including people who discontinued treatment because of side effects or any other
reason for the cycle within which discontinuation occurred) consumed twice as
much health resources as those reported for the ‘non-relapse’ group in Almond and
colleagues (2004) over 6 months. Within a year, people in the relapse model state
were assumed to consume the resources reported for the relapse group over 6
months and the resources reported for the non-relapse group over the remaining 6
months. Therefore, the annual resource use of outpatient, primary and community
care for the relapse state consisted of the 6-month resource use reported for the
relapse group (Almond et al., 2004) plus the 6-month resource use reported for the
non-relapse group. Reported resource use in Almond and colleagues (2004) was
combined with appropriate national unit costs (Curtis, 2007; Department of Health,
2008) to estimate total annual outpatient, primary and community care costs for
people in the model states of remission and relapse. The reported resource use for
the relapse and the non-relapse groups in Almond and colleagues (2004) as well as
the respective UK unit costs are presented in Table 122. Based on the above

described methods and assumptions, the annual outpatient, primary and
community care costs for the states of remission and relapse were estimated at
£5,401 and £4,323, respectively (2007 prices).
Costs associated with management of acute episodes

People experiencing an acute episode (relapse) were assumed to be treated either as
inpatients or by CRHTTs. Glover and colleagues (2006) examined the reduction in
hospital admission rates in England, following implementation of CRHTT. They
reported that the introduction of CRHTT was followed by a 22.7% reduction in
hospital admission levels. Based on this data, the economic analysis assumed that
77.3% of people with schizophrenia experiencing a relapse would be admitted to
hospital, and the remaining 22.7% would be seen by CRHTTs. However, all people
under long-term hospital care while in remission (see costs of residential care in next
subsection) were assumed to be treated as inpatients when they experienced an
acute episode.
The average cost of hospitalisation for people in acute episode was estimated by
multiplying the average duration of hospitalisation for people with schizophrenia,
schizotypal and delusional disorders (F20-F29, according to ICD-10) in England in
2006/07 (NHS The Information Centre, 2008b) by the national average unit cost per
bed-day in a mental health acute care inpatient unit for adults in 2006/07
(Department of Health, 2008).
Regarding the management of people with schizophrenia experiencing an acute

episode by CRHTTs, the GDG estimated that treatment lasted 8 weeks. This period
was multiplied by the unit cost of each case treated by CRHTTs per care staff per
week (Curtis, 2007) to provide a total cost associated with the management of acute
episodes by CRHTTs.
All people experiencing an acute episode were assumed to interrupt the

antipsychotic medication they were taking during remission and receive olanzapine
at a dose of 15mg/day (Royal College of Psychiatrists, 2008) for the duration of the
acute episode, which was assumed to be equal to the duration of hospitalisation for
people with schizophrenia (as reported by the NHS, The Information Centre, 2008a
(NHS The Information Centre, 2008b)). Olanzapine was chosen as a representative
SGA for the treatment of acute episodes; its selection was made only for modelling
purposes and does not necessarily suggest use of olanzapine instead of other
available antipsychotic drugs for the treatment of acute episodes in people with
schizophrenia.
Table 123 presents the resource use and respective unit costs associated with
management of acute episodes in people with schizophrenia, and the percentage of
people receiving each intervention.

Residential and long-term hospital care costs
The percentage of people with schizophrenia living in private households, sheltered
housing, group homes or under long-term hospital care were estimated using
respective UK data (Mangalore & Knapp, 2007). The unit costs of residential care
(sheltered housing and group homes) and long-term hospital care were taken from
national UK sources (Curtis, 2007; Department of Health, 2008). Residential and
long-term hospital care costs in the model were assumed to be independent of the
choice of antipsychotic drug and were incurred over all of the time that people were
not hospitalised for an acute episode. For this reason, the costs somewhat differed
between remission and relapse health states. Residential care costs were assumed to
be zero during management of acute episodes for those people treated as inpatients.
Long-term hospital care costs were assumed to be zero during management of acute
episodes because all people under this type of care were assumed to be treated as
inpatients once they experienced an acute episode.
The type of accommodation and the costs associated with residential and long- term
hospital care in people with schizophrenia in the economic model are reported in
Table 124.

Table 122: Resource use over 6 months and unit costs associated with outpatient, primary and community care for people with
schizophrenia
Service Mean usage per person Unit cost Sources of unit costs; comments
(Almond et al., 2004) (2007 prices)
Non-relapse Relapse
Outpatient psychiatric 1.4 2.1 £140 Department of Health, 2008a; cost per face-to-face contact in
visits outpatient mental health services
Outpatient other visits 0.1 0.3 £93 Department of Health, 2008a; cost per attendance in day care
Day hospital visits 2.3 2.1 £93 Department of Health, 2008a; cost per attendance in day care
Community mental health 2.4 1.4 £124 Department of Health, 2008a; cost per contact with CMHTs
centre visits
Day care centre visits 5.9 0.9 £93 Department of Health, 2008a; cost per attendance in day care
Group therapy 0.4 0.1 £93 Department of Health, 2008a; cost per attendance in day care
Sheltered workshop 1.1 0 £49 Curtis, 2007. Sheltered work schemes: £8.1 gross cost per hour; 6
hours per contact assumed
Specialist education 2.9 0 £93 Department of Health, 2008a; cost per attendance in day care

Other 0.6 0 £50 Assumption
(not specified)
Psychiatrist visits 2.5 2.3 £240 Department of Health, 2008a; cost per domiciliary visit by psychiatrist
Psychologist visits 0 0 £196 Department of Health, 2008a; cost per domiciliary visit by
psychologist
GP visits 1.8 1.6 £58 Curtis, 2007; cost per home visit £55 including travel, qualification
and direct care staff costs – 2006 prices
District nurse visits 0.1 0 £24 Curtis, 2007; cost per home visit for community nurse including
qualification costs and travelling
CPN visits 12.6 5.2 £26 Curtis, 2007; cost per hour of client contact for community nurse
specialist £75; assuming 20 minutes’ duration of visit; including
qualification costs and travelling
Social worker visits 0.1 0.4 £41 Curtis, 2007; cost per hour of face-to-face contact £124; assuming 20
minutes’ duration of visit – qualification costs not available
Occupational therapist 0 0.8 £39 Curtis, 2007; cost of community occupational therapist per home visit
visits including qualification and travelling costs
Home help/care worker 0.4 0.6 £19 Curtis, 2007; cost of care worker per hour of
face-to-face week day programme – qualification costs not available

Table 123: Hospital, and crisis resolution and home treatment team costs per
person in acute episode (relapse)
Treatment Duration Unit cost (2007 Total cost % of people

prices) treated
Acute hospital 111days £259/day £28,645 77.3 (Glover
(NHS, 2008a) (Department of et al., 2006)
Health, 2008a)
CRHTT 8 weeks £264 per case per £2,112 22.7 (Glover
(GDG estimate) care staff per week et al., 2006)
(Curtis, 2007)
Olanzapine 111 days £4.26/day £471 100 (assumption)

15mg/day (NHS,2008a) (BNF56)
Table 124: Type of accommodation and costs of residential and long-term hospital
care in people with schizophrenia (remission state)
Type of % of peoplea Unit cost Source of unit Weighted annual

accommodation (2007price) cost cost
Private 77 0 N/A 0
household
Residential care 18 £478/week Curtis, 2007 £4,486

(sheltered
Residential care 2 £107/week Curtis, 2007 £112
(group home)
Long-term 3 £249/day Department of £2,727

hospital care Health, 2008a
Total weighted residential cost per person in remission £7,325
Note. a Based on data reported in Mangalore & Knapp, 2007
Costs incurred by switching between antipsychotic medications

People moving to next-line treatment (because of intolerable side effects or relapse)
were assumed to incur additional costs, associated with three visits to a consultant
psychiatrist lasting 20 minutes each, at a total cost of £435 (the unit cost of a
consultant psychiatrist was £435 per hour of patient contact, including qualification
costs (Curtis, 2007)).
Costs of managing side effects and related complications

Although acute EPS may be managed solely by dose adjustment or may improve
spontaneously, people experiencing acute EPS were assumed to pay a visit to a
426
consultant psychiatrist, lasting 20 minutes, and receive procyclidine at a daily dose
of 15 mg for 3 months.
All people experiencing weight gain were assumed to pay two visits to their GP for
general advice. In addition, 20% of them received special advice from a dietician.
These methods of management were consistent with levels I and II of interventions
for people with weight gain recommended by the NICE clinical guideline on obesity
(NICE, 2006b).
Resource use estimates and respective unit costs associated with management of
acute EPS and weight gain in people with schizophrenia are reported in Table 125.
The annual cost of diabetes without complications, consisting of anti-diabetic and
antihypertensive drug treatment and inclusive of implementation costs was
estimated based on published data from UKPDS (Clarke et al., 2005). Costs
associated with management of complications from diabetes were taken from the
same study.
Costs were uplifted to 2007 prices using the Hospital and Community Health
Services Pay and Prices inflation index (Curtis, 2007). Costs and QALYs associated
with each antipsychotic treatment were discounted at an annual rate of 3.5% as
recommended by NICE (NICE, 2008a).
Table 125: Resource use and respective unit costs of managing acute EPS and
weight gain
State–event Resource use (GDG Unit costs (2007prices)

estimates)
Acute EPS
Procyclidine 5mg/day for 3 months 5mg, 28-tab = £3.35 (BNF56)
Psychiatrist 1 visit of 20 minutes Cost per hour of patient contact: £435

(qualification costs included – Curtis, 2007)
Weight gain
a
100% general 2 GP visits Cost per clinic visit:£52 (qualification and
advice direct care staff costs included – Curtis, 2007)
a
20% diet and 3 visits to dietician over 6 Cost per hour of client contact:
exercise months (duration of first visit 1 £32 (qualification costs included – Curtis,
hour; Of next 2 visits 30 2007)
minutes)
Note. a % based on GDG estimates
Table 126 reports the mean (deterministic) values of all input parameters utilised in
the economic model and provides information on the distributions assigned to
specific parameters in probabilistic sensitivity analysis.
427
Table 126: Input parameters utilised in the economic model
Input parameter Deterministic Probabilistic distribution Source of data–comments

value
Annual probability of relapse Distribution based on 10,000 mixed treatment
comparison iterations
95%credible intervals Mixed treatment comparison competing risks
Olanzapine 0.1996 0.0146 to 0.7222 model–analysis of data included in the guideline
Amisulpride 0.2988 0.0197 to 0.9042 systematic review; results for 52 weeks assumed to
Zotepine 0.1067 0.0023 to 0.5601 reflect annual probability; results for placebo
Aripiprazole 0.2742 0.0130 to 0.8531 assumed to apply to no treatment in all years
Paliperidone 0.1625 0.0025 to 0.7008 except the first year following the move to no
Risperidone 0.2761 0.0182 to 0.8785 treatment
Haloperidol 0.3317 0.0262 to 0.9028
No treatment–following years 0.4361 0.0913 to 0.8613
Flupentixol decanoate 0.2977 Beta distribution (α= 39,β= 92 according to David et al., 1999. Meta-analysis of trials
data reported in David and colleagues, 1999) comparing flupentixol decanoate versus other
depot antipsychotics; data on relapse
No treatment–first year following 0.6062 Distribution based on 10,000 mixed Mixed treatment comparison competing risks
discontinuation of treatment treatment comparison iterations – results for model–a higher probability of relapse over the first
placebo, adding the effect of abrupt 7 months (50%) was taken into account (Viguera et
discontinuation on the risk for relapse (Viguera al., 1997)
et al., 1997)
Continued

Table 126 (continued)

value
Probability of discontinuation Distribution based on10,000 mixed treatment
because of intolerable side effects– comparison iterations
first year of initiation of a particular
antipsychotic 95% credible intervals Mixed treatment comparison competing risks
Amisulpride 0.0554 0.0006 to 0.3721 systematic review; results for 52 weeks assumed to
Zotepine 0.3821 0.0120 to 0.9750 apply to the first year within initiation of a
Aripiprazole 0.1582 0.0026 to 0.7847 particular antipsychotic only
Paliperidone 0.3287 0.0039 to 0.9770
Risperidone 0.0994 0.0020 to 0.6471
Haloperidol 0.0922 0.0017 to 0.5386
Annual probability of Distribution based on 10,000 mixed

discontinuation because of other treatment comparison iterations
reasons
95%credible intervals Mixed treatment comparison competing risks
Amisulpride 0.2435 0.0139 to 0.8324 systematic review;results for 52 weeks assumed to
Zotepine 0.2253 0.0074 to 0.8189 reflect annual probability
Aripiprazole 0.3520 0.0202 to 0.9218
Paliperidone 0.3848 0.0090 to 0.9479
Risperidone 0.1761 0.0086 to 0.7141
Haloperidol 0.2516 0.0151 to 0.8290
Continued


value
Weight gain – first year of initiation Distribution based on 10,000 mixed treatment
of a particular antipsychotic comparison iterations
ORs versus haloperidol 95%credible intervals Mixed treatment comparison simple random-
Olanzapine 2.8631 1.7050 to 4.5090 effects model–analysis of data from guide line
Amisulpride 1.8604 0.7345 to 4.0360 meta-analysis of side effects;only data reported as
Aripiprazole 0.7373 0.3498 to 1.3990 ‘increase in weight ga in of ≥7% from
Paliperidone 1.0779 0.4405 to 2.1640 baseline’were considered.
Risperidone 1.0895 0.5214 to 2.0850
Zotepine 1.0895 As for risperidone
Probability of weight gain

Haloperidol 0.2000 Beta distribution (α= 31,β= 124 according to OR of zotepine versus haloperidol assumed to be
data reported in studies with time horizon up equal of that of risperidone versus haloperidol
to 12 weeks included in the guideline meta-
analysis of side effects)
Flupentixol decanoate 0.2000 As for haloperidol Extrapolation of data reported in studies with time
horizon up to 12 weeks included in the guideline
meta-analysis of side effects;only data reported
as‘increase in weight gain of ≥7% from
baseline’were considered.
Assumed to equal that for haloperidol

Continued

Input parameter Deterministic value Probabilistic distribution Source of data–comments
Acute EPS
First year of initiation of a particular Distribution based on 10,000 mixed

antipsychotic treatment comparison iterations
ORs versus haloperidol 95% credible intervals Mixed treatment comparison full random effects
Olanzapine 0.2631 0.1832 to 0.3641 model – analysis of data from guide line meta-
Amisulpride 0.3993 0.2587 to 0.5836 analysis of side effects; only data on
Zotepine 0.1476 0.0517 to 0.3132 ‘need for anticholinergic medication’ were
Aripiprazole 0.2517 0.1505 to 0.4002 considered
Paliperidone 0.2983 0.1179 to 0.6214
Risperidone 0.4743 0.3680 to 0.5994
Probability of acute EPS

Haloperidol 0.5367 Beta distribution (α = 928, β = 801 Extrapolation of data reported in studies with time
according to data reported in RCTs with horizon up to 8 weeks included in the guideline
time horizon up to 8 weeks included in the meta-analysis of side effects; only data on ‘need for
guideline meta- analysis of side effects) anticholinergic medication’ were considered
Flupentixol decanoate 0.4891 Beta distribution (α = 45, β = 47 according David et al., 1999. Meta-analysis of trials comparing
to data reported in David and colleagues, flupentixol decanoate versus other depot
1999) antipsychotics; data on need for anti cholinergic
medication
Following years
Probability of acute EPS
All antipsychotics 10% of first year N/A (no distribution assigned) GDG expert opinion
estimate
Continued

Probability of diabetes–first year Distribution based on 10,000 mixed Probability of haloperidol estimated from
of initiation of a particular treatment comparison iterations of data on data reported in van Winkel et al., 2006
antipsychotic weight gain and 2008 and considering the increased RR
Olanzapine 0.0417 Relative risk of each SGA versus haloperidol for diabetes of SGAs versus FGAs; the
Amisulpride 0.0317 for diabetes was assumed to equal their in- remaining probabilities were calculated by
Zotepine 0.0214 between relative risk for weight gain; the multiplying respective RRs for weight gain
Aripiprazole 0.0156 latter was deter-mined by the posterior of each SGA versus haloperidol by the
Paliperidone 0.0212 distribution of ORs of weight gain for each probability of diabetes for haloperidol
Risperidone 0.0214 SGA and haloperidol
Haloperidol 0.0200 Beta distribution (α= 2, β= 98 based on

assumption)
Flupentixol decanoate 0.0200 As for haloperidol
Continued

Probability of glucose Distribution based on 10,000 mixed Probability of haloperidol estimated from
intolerance– first year of treatment comparison iterations of data on data identified in the guideline systematic
initiation of a particular weight gain review; the remaining probabilities were
antipsychotic calculated by multiplying respective RRs
Olanzapine 0.3129 Relative risk of each SGA versus haloperidol for weight gain of each SGA versus
Amisulpride 0.2381 for glucose intolerance was assumed to equal haloperidol by the probability of glucose
Zotepine 0.1606 their in-between relative risk for weight gain; intolerance for haloperidol
Aripiprazole 0.1167 the latter was determined by the posterior
Paliperidone 0.1592 distribution of ORs of weight gain
Risperidone 0.1606 For each SGA and haloperidol, respectively
Haloperidol 0.1500
Beta distribution (α= 15, β= 85 based on
assumption)
Flupentixol decanoate 0.1500
As for haloperidol
Annual transition probability of 0.0196 Beta distribution Gillies et al., 2008
impaired glucose tolerance to Standard error 0.0025 (Gillies et al., 2008)
diabetes
Continued

Annual probability of
diabetes complications
Fatal myocardial infarction 0.0042 Beta distribution Based on UKPDS data (Stratton et al., 2000),
Non-fatal myocardial 0.0130 Determined from the numbers of assuming that 20% of people with schizophre-
infarction 0.0039 people experiencing each of the nia and diabetes in the model had Hgb A1C
Non-fatal stroke 0.0023 complications at each level of Hgb concentration 7 to <8%, 30% of people had 8 to
Amputation 0.0040 A1C concentration in the UKPDS <9%, 30% of people had 9 to <10% and 20% of
Macrovascular events – heart 0.0157 (Stratton et al., 2000) people had ≥10%
failure Microvascular events
– ischaemic heart disease
Standardised mortality ratio 2.6 N/A (no distribution assigned) McGrath et al., 2008
– all cause mortality
Mortality rates per 1000 25–34 years: 0.69 N/A (no distribution assigned) Office for National Statistics, 2008; mortality
people in general 35–44 years: 1.29 rates for England and Wales, 2005, estimated
population by age 45–54 years: 3.10 based on a male to female ratio 1.4 to 1,
55–64 years: 7.53 characterising people with schizophrenia
65–74 years: 20.48 (McGrath, 2006)
75–84 years: 59.36
≥85 years: 164.02
Utility scores
Model health states Beta distribution Lenert et al., 2004; linking between model states
Remission 0.799 Determined using the reported and states described in the study based on GDG
Relapse 0.670 numbers of people valuing each estimates – see the main text for details.
Death 0.000 PANSS-generated health state as in Duration of decrement in HRQoL caused by
Lenert and colleagues (2004) relapse: 6 months


value
Side effects
Acute EPS –0.888% Estimated from the number o fpeople Lenert et al., 2004; acute EPS causes
Weight gain –0.959% valuing the presence of each side effect, HRQoL reduction corresponding to that
as reported in Lenert and colleagues of pseudo-parkinsonism, lasting 3
(2004) months; weight gain causes permanent
reduction in HRQoL
Diabetes complications 95% credible intervals
Myocardial infarction –0.055 –0.067 to –0.042 Clarke et al., 2002; utility scores based on
Stroke –0.164 –0.222 to –0.105 patient-reported EQ-5D scores, valued
Amputation –0.280 –0.389 to –0.170 using EQ-5DUK tariff values
Macrovascular events – heartfailure –0.108 –0.169 to –0.048
Microvascular events – ischaemic heart disease –0.090 –0.126 to –0.054
Annual drug acquisition costs N/A (no distribution assigned) BNF56 (British Medical Association &
(remission state) the Royal Pharmaceutical Society of
Olanzapine £1,036 Great Britain, 2008), except risperidone
Amisulpride £696 cost, which was taken from the
Zotepine £767 Electronic Drug Tariff (NHS, Business
Aripiprazole £1,325 Services Authority, 2008). Average daily
Paliperidone £1,902 dosage taken from respective NHS data
Risperidone £821 (NHS, The Information Centre,
Haloperidol £175 2008c) and BNF guidance when no other
Flupentixol decanoate £81 data were available
Continued


value
Annual costs of remission Gamma distribution
Outpatient, primary and community care £5,401 Standard error of all costs: 70% of mean Details on outpatient, primary and
Residential and long-term hospital care £7,325 value (assumption) community care cost reported in Table
Total (cost of antipsychotic medication for £12,726 122; details on costs of residential and
relapse prevention excluded) long-term hospital care reported in Table
124; 2007 prices
Annual costs of relapse Gamma distribution Details on outpatient, primary and

Outpatient, primary and community care £4,323 Standard error of all costs: 70% of mean community care cost reported in Table
Residential and long-term hospital care Acute £5,421 value (assumption) 122; details on costs of treating acute
treatment (including olanzapine) Total (cost of £23,274 episode reported in Table 123; details on
antipsychotic medication for relapse prevention £33,018 costs of residential and long-term
excluded) hospital care reported in Table 124; 2007
prices
Cost of switching between antipsychotics £435 3 visits to consultant psychiatrist, lasting
Standard error: 70% of mean value 20 minutes each; unit cost from Curtis,
(assumption) 2007; 2007 prices
Continued


value
Cost of treating side effects Gamma distribution Details on resource use and unit costs
Acute EPS £177 Standard error of all costs: 70% of the associated with acute EPS and weight
Weight gain £117 respective mean value (assumption) gain reported in Table 125; 2007 prices
Diabetes (without
complications) – annual £199 UKPDS (Clarke et al., 2005); 2007 prices
Fatal myocardial infarction £1,531
Non-fatal myocardial infarction
first year/following years £5,407/£616
Non-fatal stroke
Amputation
Macrovascular events-heart failure
first year/following years £418/£343
Microvascular events-ischaemic heart disease
first year/following years £363/£271
Discount rate (for both costs and outcomes) 0.035 N/A (no distribution assigned) Recommended by NICE (NICE, 2008a)

11.2.10 Data analysis and presentation of the results
Two methods were employed to analyse the input parameter data and present the
results of the economic analysis.
First, a ‘deterministic’ analysis was undertaken, where data are analysed as point
estimates; results are presented as mean total costs and QALYs associated with each
treatment option are assessed. Relative cost effectiveness between alternative
treatment options is estimated using incremental analysis: all options are initially
ranked from most to least effective; any options that are more expensive than
options that are ranked higher are dominated (because they are also less effective)
and excluded from further analysis. Subsequently, ICERs are calculated for all pairs
of consecutive options. ICERs express the additional cost per additional unit of
benefit associated with one treatment option relative to its comparator. Estimation of
such a ratio allows consideration of whether the additional benefit is worth the
additional cost when choosing one treatment option over another.
If the ICER for a given option is higher than the ICER calculated for the previous
intervention in ranking, then this strategy is also excluded from further analysis, on
the basis of extended dominance. After excluding cases of extended dominance,
ICERs are recalculated. The treatment option with the highest ICER below the cost
effectiveness threshold is the most cost-effective option.
A number of sensitivity analyses explored the impact of the uncertainty

characterising model input parameters on the results of the deterministic analysis.
The following scenarios were tested:
• Unit cost per bed-day in an adult mental health acute care inpatient
unit of £235, according to the reported lower quartile of the NHS
reference unit cost (Department of Health, 2008)
• Duration of hospitalisation for people experiencing an acute episode of
69 days, taken from an effectiveness trial of clozapine versus SGAs
conducted in the UK (CUtLASS Band 2, (Davies et al., 2008)
• Combination of the two scenarios above.
The following three scenarios attempted to investigate the impact of hospitalisation

costs on the results of the analysis:
• Use of alternative utility scores for schizophrenia health states, as
reported in Chouinard and Albright (1997) and Glennie (1997)
• Probability of side effects assumed to be common for all antipsychotic
drugs: probabilities of acute EPS, weight gain and, subsequently,
glucose intolerance and diabetes were assumed to be the same for all
drugs. This scenario aimed at exploring the importance of side effects
in determining total QALYs, costs and relative cost effectiveness
between antipsychotic medications over time
• Probability of relapse assumed to be common for all antipsychotic
drugs. The objective of this sensitivity analysis was to explore whether

the effectiveness in preventing relapse was the driver of the cost
effectiveness results, as expected.
In addition to deterministic analysis, a ‘probabilistic’ analysis was also conducted. In

this case, most of the model input-parameters were assigned probability
distributions (rather than being expressed as point estimates), to reflect the
uncertainty characterising the available clinical and cost data. Subsequently, 10,000
iterations were performed, each drawing random values out of the distributions
fitted onto the model input parameters. This exercise provided more accurate
estimates of mean costs and benefits for each antipsychotic (averaging results from
the 10,000 iterations) by capturing the non- linearity characterising the economic
model structure (Briggs et al., 2006a).
The probabilistic distributions of data on relapse, discontinuation and side effects

that were analysed using mixed treatment comparison techniques (that is, annual
probability of relapse, probability of treatment discontinuation because of intolerable
side effects and annual probability of treatment discontinuation because of any other
reason, ORs of weight gain versus haloperidol and ORs of acute EPS versus
haloperidol) were defined directly from random values recorded for each of the
10,000 respective mixed treatment comparison iterations performed in Winbugs. To
maintain the correlation between the posterior estimates for (i) probability of relapse,
(ii) probability of treatment discontinuation because of intolerable side effects and
(iii) probability of treatment discontinuation because of any other reason, data from
each of the common mixed treatment comparison simulations for these parameters
were exported jointly and fitted into the Excel file of the economic model where the
probabilistic analysis was carried out.
The probability of relapse and acute EPS for the depot antipsychotic, and of acute
EPS and weight gain for haloperidol, were given a beta distribution. Beta
distributions were also assigned to utility scores and rates of complications from
diabetes. The estimation of distribution ranges in all these cases was based on
available data in the published sources of evidence or from the guideline meta-
analysis.
The probabilities of developing diabetes and glucose impairment following use of

haloperidol were also given a beta distribution; the ranges of values attached to
these parameters were based on assumptions.
All costs (except drug acquisition costs) were assigned a gamma distribution; to take
account of their likely high skewness and variability, the standard errors associated
with costs were assumed to equal 70% of the values used in deterministic analysis.
Table 126 shows which input parameters were assigned distributions in the
probabilistic analysis, and gives more details on the types of distributions and the
methods employed to define their range.

Results of probabilistic analysis are presented in the form of cost-effectiveness
acceptability curves (CEACs), which demonstrate the probability of each treatment
option being the most cost effective among the strategies assessed at different levels
of willingness-to-pay per unit of effectiveness (that is, at different cost-effectiveness
thresholds the decision-maker may set). In addition, the cost effectiveness
acceptability frontier (CEAF) is provided alongside CEACs, showing which
treatment option among those examined offers the highest average net monetary
benefit (NMB) at each level of willingness-to-pay (Fenwick et al., 2001). The NMB of
a treatment option at different levels of willingness-to-pay is defined by the
following formula:
NMB = E . λ - C
where E and C are the effectiveness (number of QALYs) and costs associated with
the treatment option, respectively, and λ is the level of the willingness-to-pay per
unit of effectiveness.
11.3 RESULTS
11.3.1 Results of deterministic analysis
According to deterministic analysis, zotepine was the most cost-effective option
among those assessed because it produced the highest number of QALYs and was
associated with the lowest costs (dominant option). This result was observed for
both time horizons of the analysis; that is, 10 years and lifetime.
Table 127 provides mean costs and QALYs for every antipsychotic drug assessed in
the economic analysis, as well as the results of incremental analysis, over a time
horizon of 10 years. The seven drugs have been ranked from the most to the least
effective in terms of number of QALYs gained. Zotepine is associated with lowest
costs and highest benefits (QALYs) and consequently dominates all other treatment
options. It can be seen that paliperidone and olanzapine dominate all drugs except
zotepine; therefore, if zotepine is not an option for the treatment of people with
schizophrenia that is in remission, then the decision (solely in terms of cost
effectiveness) would have to be made between paliperidone and olanzapine. The
ICER of paliperidone versus olanzapine is £150,159/QALY; this figure is much
higher than the cost effectiveness threshold of £20,000–£30,000/QALY set by NICE
(NICE, 2008b). Therefore, at 10 years of antipsychotic medication use, according to
the results of deterministic analysis, olanzapine is the second most cost-effective
option following zotepine, and paliperidone is the third (because it dominates all
other options). If paliperidone and olanzapine are excluded from analysis (in
addition to zotepine), then four drugs remain for further analysis: two of them,
aripiprazole and amisulpride, are dominated by haloperidol. The ICER of
risperidone to haloperidol exceeds £1,600,000/QALY, and therefore haloperidol is
the most cost-effective option among the four remaining drugs.

Table 127: Mean costs and QALYs per person for each antipsychotic drug used for relapse prevention in people with
schizophrenia that is in remission – time horizon of 10 years. Incremental analysis undertaken in steps, after excluding the
most cost-effective option of the previous step, to enable ranking of medications in terms of cost effectiveness
Antipsychotic drug QALYs Cost Incremental analysis (cost per QALY gained)
All options Excluding Excluding Excluding Excluding
zotepine and paliperidone haloperidol aripiprazole
olanzapine
Zotepine 6.468 £139,170 Dominant

Paliperidone 6.427 £142,173 Dominated £150,159
Olanzapine 6.420 £141,212 Dominated
Risperidone 6.417 £149,112 Dominated Dominated £1,600,986 £204,529 £48,961
Haloperidol 6.413 £143,406 Dominated Dominated
Aripiprazole 6.400 £145,697 Dominated Dominated Dominated
Amisulpride 6.392 £147,920 Dominated Dominated Dominated Dominated

By repeating this process in steps, and excluding in each new incremental analysis
all options found to be cost effective in previous ones, it is possible to rank all
medications in terms of cost effectiveness. This incremental analysis ‘in steps’
resulted in the following ranking of antipsychotics in terms of cost effectiveness: (1)
zotepine; (2) olanzapine; (3) paliperidone; (4) haloperidol; (5) aripiprazole; (6)
amisulpride; (7) risperidone.
Table 128 provides mean costs and QALYs for each antipsychotic drug assessed in
the economic model as well as results of incremental analysis in steps over a lifetime.
The seven drugs have again been ranked from the most to the least effective.
Zotepine dominates all other options in this analysis, too. If zotepine is excluded
from the analysis, then paliperidone dominates all other drugs except haloperidol
and olanzapine. The ICER of paliperidone versus haloperidol is £11,458 per QALY;
the ICER of haloperidol versus olanzapine is £41,129 per QALY. Consequently,
haloperidol is excluded from consideration on the basis of extended dominance. The
ICER of paliperidone versus olanzapine is £20,872 per QALY. These figures suggest
that, if zotepine is not an option, then olanzapine is the second best option in terms
of cost effectiveness (using the lower, £20,000/QALY, threshold set by NICE
(2008b)), and paliperidone third (however, it must be noted that the figure of
£20,872/QALY is very close to the lower threshold and if the upper NICE cost
effectiveness threshold of £30,000/QALY is used, then paliperidone is ranked second
best option in terms of cost effectiveness and olanzapine third). If incremental
analysis in steps is undertaken, as show Table 128, then the ranking of antipsychotic
medications in terms of cost effectiveness is the following: (1) zotepine; (2)
olanzapine; (3) paliperidone; (4) haloperidol; (5) aripiprazole; (6) risperidone; (7)
amisulpride.
A comparison of rankings in terms of QALYs between Table 127 and Table 128
shows that olanzapine and haloperidol appear in low places in the lifetime horizon
(seventh and fifth, respectively), compared with their ranking at 10 years where they
are ranked third and fourth, respectively. This finding is explained by the higher risk
for weight gain and diabetes characterising olanzapine (olanzapine was the second-
line antipsychotic in the cohort initiated on haloperidol); eventually, the (permanent)
increase in weight and the incidence of complications from diabetes, which was
higher in the cohorts receiving olanzapine as first or second-line treatment, reduced
the overall HRQoL and the total number of QALYs gained relative to other
treatment options. Nonetheless, the ranking of olanzapine and haloperidol in terms
of cost effectiveness was not affected: they were ranked second and fourth cost-
effective options, respectively, over 10 years, and this ranking order remained over a
lifetime. It must be noted that, with the exception of the last two places, the ranking
of antipsychotic medications in terms of cost effectiveness was not affected by the
time horizon used.

Table 128: Mean costs and QALYs per person for each antipsychotic drug used for relapse prevention in people with
schizophrenia that is in remission – lifetime horizon. Incremental analysis undertaken in steps, after excluding the most cost-
effective option of the previous step, to enable ranking of medications by cost effectiveness
Antipsychotic QALYs Cost Incremental analysis (cost per QALY gained)

drug
All options Excluding Excluding Excluding Excluding Excluding
zotepine olanzapine paliperidone haloperidol aripiprazole
Zotepine 16.849 £397,247 Dominant
Paliperidone 16.804 £402,288 Dominated £20,872 £11,458
Risperidone 16.791 £409,083 Dominated Dominated Dominated £191,056 £118,464 £12,809
Aripiprazole 16.767 £406,195 Dominated Dominated Dominated Ext.domin.
Haloperidol 16.753 £401,702 Dominated Ext.domin.
Amisulpride 16.733 £408,332 Dominated Dominated Dominated Dominated Dominated
Olanzapine 16.729 £400,725 Dominated
Note. Ext.domin. = extendedly dominated.

Figure 6 and Figure 7 present the cost effectiveness planes for the two time horizons
of the analysis, showing the incremental costs and benefits (QALYs) of all SGAs
versus haloperidol. In both cases, it can be seen that zotepine is in the southeast
quadrant and has the highest number of QALYs and the lowest costs relative to all
other options assessed.
Figure 6: Cost-effectiveness plane of all treatment options plotted against

haloperidol, at 10 years of antipsychotic medication use
Cost effectiveness plane - 10 years

£6,000
Difference in costs
£4,000
Olanzapine
£2,000 Amisulpride
Zotepine
£0
-0.04 -0.02 0.00 0.02 0.04 0.06 Aripiprazol
Paliperidone
-£2,000
Risperidone
-£4,000 Haloperidol
-£6,000 Difference in QALYs

Figure 7: Cost-effectiveness plane of all treatment options plotted against
haloperidol, over a lifetime of antipsychotic medication use
Cost effectiveness plane - over life time

£8,000
£6,000
Difference in costs
£4,000
Olanzapine
£2,000 Amisulpride
Zotepine
£0
-0.04 -0.02 0.00 0.02 0.04 0.06 0.08 0.10 Aripiprazol
Paliperidone
-£2,000
Risperidone
-£4,000 Haloperidol
-£6,000
Difference in QALYs
Results of deterministic sensitivity analysis

Results were very sensitive to annual probabilities of relapse, as expected. When all
antipsychotic medications were assumed to have equal probabilities of relapse, the
ranking of medications in terms of effectiveness was significantly affected. In
general, this ranking by effectiveness was predicted by the ranking of medications in
terms of discontinuation to other reasons, with options with lower probabilities of
discontinuation ranking more highly in terms of effectiveness. Regarding cost
effectiveness, the ranking of treatment options at 10 years following incremental
analysis in steps was: (1) haloperidol; (2) amisulpride; (3) olanzapine; (4)
aripiprazole; (5) risperidone; (6) zotepine; (7) paliperidone. Over a lifetime, the
ranking of antipsychotic medications in terms of cost effectiveness was: (1)
risperidone; (2) amisulpride; (3) haloperidol; (4) olanzapine; (5) aripiprazole; (6)
zotepine; (7) paliperidone. It is obvious that results were greatly affected by this
scenario, with options that were ranked highly in base-case deterministic analysis,
such as zotepine and paliperidone, occupying the last two places in ranking when
relapse rates were assumed to be the same for all treatment options.
Results were, overall, robust under the other scenarios explored in sensitivity
analysis. In all cases, zotepine was the most cost-effective option: zotepine remained
dominant under all other hypotheses tested, with the exception of the scenario that

combined a low estimate of inpatient stay for people having an acute episode (69
days instead of 111, which was the estimate used in base-case analysis) with a lower
respective unit cost. In this case, and over a time horizon of 10 years, zotepine
dominated all treatment except olanzapine which became less costly. However, the
ICER of zotepine versus olanzapine was £7,751/QALY; therefore, zotepine remained
the most cost-effective option of those assessed.
Ranking of medications in terms of cost effectiveness did not change at 10 years

under any scenario of those examined (with the exception of using common
probabilities of relapse, as discussed above). However, over a lifetime, some of the
tested scenarios did affect the ranking of antipsychotic medications. Table 129
provides the ranking of medications in terms of cost effectiveness for those scenarios
that affected ranking over a lifetime (the scenario of using common probabilities of
relapse has not been presented in this table, as it has been discussed above).
Table 129: Ranking of antipsychotic medications in terms of cost effectiveness

over a lifetime under: (1) base-case analysis; (2) use of a lower estimate of
inpatient stay; (3) use of a lower estimate of inpatient stay and a lower unit cost of
mental health inpatient bed-day; (4) use of utility scores reported in Glennie
(1997); (5) assumption of common probabilities of side effects for all antipsychotic
medications
Base-case analysis Scenario tested in sensitivity analysis

1 2 3 4 5
Zotepine Zotepine Zotepine Zotepine Zotepine
Olanzapine Paliperidone Paliperidone Paliperidone Olanzapine
Paliperidone Olanzapine Haloperidol Olanzapine Haloperidol
Haloperidol Haloperidol Olanzapine Haloperidol Paliperidone
Aripiprazole Aripiprazole Aripiprazole Aripiprazole Aripiprazole
Risperidone Amisulpride Amisulpride Risperidone Amisulpride
Amisulpride Risperidone Risperidone Amisulpride Risperidone
It must be noted that using common probabilities of side effects (that is, acute EPS,
weight gain, glucose intolerance and diabetes) for all antipsychotic medications did
not significantly affect the results of the analysis. Ranking medications in terms of
QALYs changed, as expected, with olanzapine being ranked in second place in both
of the time horizons examined. However, the first two ranked places in terms of cost
effectiveness were not affected, with zotepine remaining the most cost-effective
option followed by olanzapine, as in base-case analysis.
11.3.2 Results of probabilistic analysis

Results of probabilistic analysis did not differ significantly from those of
deterministic analysis: as in deterministic analysis, zotepine dominated all other
options because it was associated with the lowest total costs and highest total

QALYs (that is, mean values from 10,000 iterations) compared with the other six
antipsychotic medications assessed. Regarding the ranking of medications in order
of cost effectiveness, this was the same for deterministic and probabilistic analysis
over 10 years. Over a lifetime, cost-effectiveness ranking of antipsychotic drugs in
probabilistic analysis differed from respective ranking in deterministic analysis to
some extent; probabilistic analysis ranking was as follows: (1) zotepine; (2)
olanzapine; (3) haloperidol; (4) paliperidone; (5) risperidone; (6) amisulpride; (7)
aripiprazole.
Probabilistic analysis demonstrated that zotepine had the highest probability of

being the most cost-effective option among all antipsychotic medications examined,
at any level of willingness-to-pay per additional QALY gained of those explored;
that is, from zero to £50,000 per QALY gained. However, this probability was low,
ranging between 25 and 29% at 10 years, and 28 and 33% over a lifetime, and
remained virtually unaffected by the cost-effectiveness threshold examined. The
other antipsychotic medications had probabilities of being the most cost-effective
options that ranged from approximately 5% (haloperidol) to 16% (paliperidone) and
were also almost independent of the cost-effectiveness threshold and the time
horizon examined. The cost effectiveness acceptability frontier coincided with the
CEAC for zotepine, because zotepine produced the highest average net benefit at
any level of willingness to pay.
Figure 8 and Figure 9 show the CEACs generated for each of the seven antipsychotic
medications examined, over 10 years and a lifetime of antipsychotic medication use,
respectively.
Table 130 and Table 131 show the probabilities of each antipsychotic medication
being cost effective at various levels of willingness-to-pay per QALY gained.

Figure 8: Cost-effectiveness acceptability curves of all treatment options at 10
years of antipsychotic medication use
Cost effectiveness acceptability

curves: use of antipsychotic
medication for relapse prevention - 10
years
0.30
Probability of cost effectiveness
0.25
Olanzapine
0.20
Amisulpride
0.15 Zotepine
Aripiprazole
0.10
Paliperidone
0.05
Risperidone
0.00 Haloperidol
£0 £10,000 £20,000 £30,000 £40,000 £50,000
Willingness-to-pay per QALY gained
Figure 9: Cost-effectiveness acceptability curves of all treatment options over a

lifetime of antipsychotic medication use
Cost effectiveness acceptability

curves: use of antipsychotic
medication for relapse prevention -
over lifetime
0.35
Probability of cost effectiveness
0.30
0.25
Olanzapine
Amisulpride
0.20
Zotepine
0.15
Aripiprazole
0.10
Paliperidone
0.05 Risperidone
0.00 Haloperidol
£0 £10,000 £20,000 £30,000 £40,000 £50,000
Willingness-to-pay per QALY gained

11.4 DISCUSSION OF FINDINGS - LIMITATIONS OF THE
ANALYSIS
The results of the economic analysis suggest that zotepine is potentially the most
cost-effective pharmacological treatment of those examined for relapse prevention in
people with schizophrenia that is in remission. Zotepine dominated all other
treatment options in deterministic analysis. In probabilistic analysis, use of zotepine
yielded the maximum average net benefit and demonstrated the highest probability
of being the most cost-effective option at any level of willingness-to-pay per unit of
effectiveness. However, because of the high uncertainty characterising model input
parameters, the probability of zotepine being the most cost-effective option was low
at approximately 27 to 30% and remained virtually unaffected by the level of
willingness-to-pay. The probability of zotepine being the most cost-effective
antipsychotic medication at the NICE cost-effectiveness threshold of £20,000 per
QALY was 27.17% at 10 years and 30.46% over a lifetime.
One of the major drawbacks of the economic analysis was the omission of a number
of antipsychotic drugs that are potentially effective in preventing relapse in people
with schizophrenia in remission. Quetiapine and FGAs other than haloperidol were
not assessed in the economic analysis because no relevant clinical data in the area of
relapse prevention were identified in the systematic review of relevant literature.
The clinical data on relapse and discontinuation utilised in the economic model were
limited in some cases: data on zotepine, which was shown to be the dominant option
in deterministic analysis, were derived exclusively from a placebo-controlled RCT.
Respective data on aripiprazole and paliperidone were also taken from two trials
that assessed each of these two antipsychotic drugs versus placebo. Therefore, the
results of the economic analysis should be interpreted with caution.

Table 130: Probability of each antipsychotic intervention being cost effective at various levels of willingness-to-pay per QALY
gained (WTP) – 10 years
WTP Olanzapine Amisulpride Zotepine Aripiprazole Paliperidone Risperidone Haloperidol

0 0.1457 0.1363 0.2552 0.1492 0.1736 0.0911 0.0489
£5,000 0.1436 0.1364 0.2582 0.1466 0.1726 0.0939 0.0487
£10,000 0.1427 0.1357 0.2633 0.1442 0.1710 0.0955 0.0476
£15,000 0.1410 0.1364 0.2675 0.1420 0.1686 0.0967 0.0478
£20,000 0.1407 0.1341 0.2717 0.1413 0.1666 0.0982 0.0474
£25,000 0.1404 0.1341 0.2757 0.1387 0.1641 0.0998 0.0472
£30,000 0.1390 0.1338 0.2795 0.1370 0.1626 0.1014 0.0467
£35,000 0.1389 0.1333 0.2806 0.1357 0.1607 0.1034 0.0474
£40,000 0.1381 0.1324 0.2835 0.1343 0.1586 0.1054 0.0477
£45,000 0.1377 0.1322 0.2861 0.1323 0.1566 0.1072 0.0479
£50,000 0.1369 0.1312 0.2887 0.1301 0.1553 0.1092 0.0486

Table 131: Probability of each antipsychotic intervention being cost effective at various levels of willingness-to-pay per QALY
gained (WTP) – over a lifetime
WTP Olanzapine Amisulpride Zotepine Aripiprazole Paliperidone Risperidone Haloperidol

0 0.1412 0.1440 0.2801 0.1216 0.1476 0.1172 0.0483
£5,000 0.1294 0.1402 0.2863 0.1213 0.1488 0.1218 0.0522
£10,000 0.1218 0.1381 0.2924 0.1203 0.1484 0.1257 0.0533
£15,000 0.1143 0.1363 0.2984 0.1196 0.1483 0.1289 0.0542
£20,000 0.1060 0.1349 0.3046 0.1171 0.1485 0.1331 0.0558
£25,000 0.1007 0.1340 0.3092 0.1161 0.1464 0.1364 0.0572
£30,000 0.0960 0.1316 0.3140 0.1146 0.1471 0.1399 0.0568
£35,000 0.0921 0.1288 0.3182 0.1145 0.1472 0.1425 0.0567
£40,000 0.0882 0.1281 0.3224 0.1125 0.1458 0.1461 0.0569
£45,000 0.0853 0.1260 0.3261 0.1109 0.1449 0.1497 0.0571
£50,000 0.0831 0.1245 0.3279 0.1100 0.1443 0.1531 0.0571

Moreover, definition of relapse varied across the 17 trials that provided data on
relapse; this is another factor that should be taken into account when interpreting the
economic findings. Data on relapse, discontinuation because of side effects and
discontinuation because of other reasons were treated as mutually exclusive in
analysis. Although the majority of the 17 RCTs that formed the evidence-base for the
economic analysis reported these outcomes as such (that is, trial participants could
either stay in remission, or relapse, or discontinue because of side effects, or
discontinue because of other reasons), a small number of trials did not clarify
whether some participants could have been double-counted in the reporting of
outcomes and an assumption of mutual exclusiveness of such outcomes also in these
studies had to be made. Results of the mixed treatment comparison analysis of
clinical data on relapse prevention were characterised by high uncertainty, as
demonstrated by the wide 95% credible intervals of the respective posterior
distributions; this uncertainty was reflected in the results of the probabilistic
economic analysis: the probability of zotepine being the most cost-effective option
was roughly 27 to 30%, with the probabilities of the remaining options being cost
effective ranging from around 5% (haloperidol) to 16% (paliperidone), regardless of
the level of willingness-to-pay per QALY gained.
The mixed treatment comparison analysis of the available clinical data, including
relapse and discontinuation rates as well as rates of side effects, overcame the major
limitation characterising previous economic models that assessed the cost
effectiveness of pharmacological treatments for people with schizophrenia: most of
those analyses synthesised trial-based evidence by naive addition of clinical data
across relevant treatment arms, thus breaking randomisation rules and introducing
bias into the analysis (Glenny et al., 2005). On the other hand, mixed treatment
comparison techniques enable evidence synthesis from both direct and indirect
comparisons between treatments, and allow simultaneous inference on all
treatments examined in pair-wise trial comparisons while respecting randomisation
(Caldwell et al., 2005; Lu & Ades, 2004).
The guideline economic analysis, in contrast to previous economic studies,

considered a lifetime horizon (in addition to a time horizon of 10 years); this was
deemed appropriate and relevant for the economic question, given the potential
need for long-term (likely to be over a lifetime) use of antipsychotic drugs by people
with schizophrenia in remission, and the nature of schizophrenia, which is often
characterised by phases of remission alternating with phases of relapse over a
lifetime. However, one limitation of the analysis was the extrapolation of relatively
short-term clinical data over a lifetime because no appropriate long-term data were
available to inform the economic model: clinical data on relapse and discontinuation
were taken from trials with time horizons ranging between 26 and 104 weeks. The
52-week probability of relapse, the 52-week probability of treatment discontinuation
because of intolerable side effects and the 52-week probability of treatment
discontinuation because of any other reason were estimated in most cases by
extrapolating the available clinical data; the estimated probability of relapse and of
treatment discontinuation because of other reasons were then assumed to apply to
every yearly cycle in the model, over a lifetime of the hypothetical study cohorts.
Although such an extrapolation of the data was required to populate the economic
model, no robust evidence exists to confirm that such extrapolation accurately
reflects the long-term effectiveness of antipsychotic medication and its impact on the
course of schizophrenia in real life. If the effectiveness of antipsychotic drugs in
preventing relapse is maintained over time, then the results of the economic analysis
more closely reflect a realistic situation. If, however, the effectiveness of
antipsychotic drugs in preventing relapse is reduced over time, then this analysis
has overestimated the cost effectiveness of antipsychotic medication, especially of
those treatments that have been demonstrated to be the most effective in preventing
relapse in the short term, such as zotepine.
The economic model structure incorporated three side effects: acute EPS, weight
gain, and diabetes/glucose intolerance potentially leading to diabetes. The choice of
side effects was based on their expected impact on the relative cost effectiveness of
antipsychotic medications and the availability of relevant data. However, it should
be emphasised that antipsychotic drugs are characterised overall by a wider range of
side effects, such as other neurologic side effects including tardive dyskinesia, sexual
dysfunction, increase in prolactin levels, as well as cardiovascular and
gastrointestinal side effects, the omission of which may have affected the results of
the economic analysis. In particular, lack of consideration of tardive dyskinesia,
which has lasting effects and causes a significant impairment in HRQoL, is
acknowledged as a limitation of the analysis. Inclusion of tardive dyskinesia in the
model structure might disfavour haloperidol, given that clinical evidence indicates
that haloperidol is associated with a higher risk for neurologic side effects.
To populate the economic model using the available data on side effects, a number
of GDG estimates and further assumptions were required, including selection of
data for analysis and extrapolation of available evidence over the time horizon of the
analysis. Data on acute EPS were more comprehensive compared with data on
weight gain and data on the risk for diabetes and glucose intolerance. Data on
weight gain were not available for zotepine; for this reason the risk of weight gain
for zotepine was assumed to be equal to the respective risk for risperidone. Data on
the risk for diabetes and glucose intolerance associated with antipsychotic
medication and appropriate for the economic analysis were very sparse and not
available for all drugs assessed in the analysis. However, these parameters were
considered to be important for inclusion in the model structure, as use of
antipsychotic medication is associated with increased risk for development of
diabetes, the complications of which have been shown to affect quality of life
considerably and to incur substantial costs in the long term; therefore, to explore the
impact of such parameters on the relative cost effectiveness of antipsychotic
medications over time, a number of assumptions were made. It is acknowledged that
the estimates used in the model regarding diabetes and glucose intolerance could be
potentially conservative and may not fully reflect the negative effect of antipsychotic
medication on glucose metabolism.
Deterministic analysis showed that although olanzapine was ranked second in terms
of effectiveness (number of QALYs gained) at 10 years of antipsychotic medication
use, it was placed last in the ranking when a lifetime horizon was considered. This
change in ranking over time was probably caused by the eventual impairment in
HRQoL of people taking olanzapine, owing to the estimated higher levels of
permanent weight increase and the frequent presence of complications because of
diabetes associated with use of olanzapine compared with other antipsychotic
medications. Nevertheless, despite being the least effective option over a lifetime,
olanzapine was still ranked second in terms of cost effectiveness among the
antipsychotic drugs assessed in deterministic analysis. It must be emphasised that
deterministic sensitivity analysis revealed that the probabilities of side effects used
in the economic model had no significant impact on the overall conclusions of the
incremental analysis, because assuming equal probabilities for side effects for all
medications did not change their ranking in terms of cost effectiveness at 10 years
and led to minor changes in ranking over a lifetime (zotepine and olanzapine were
still ranked first and second most cost-effective options, respectively). However, if
the estimates used in the model regarding diabetes and glucose intolerance are
conservative and do not fully capture the negative impact of antipsychotic
medication on HRQoL and associated costs, then the relative cost effectiveness of
drugs with more significant metabolic implications, such as olanzapine, may have
been overestimated.
Data on treatment discontinuation because of intolerable side effects and side- effect
data were analysed separately. In probabilistic economic analysis, the probability of
treatment discontinuation because of intolerable side effects was varied
independently from the probability of developing each of the three side effects
examined. However, there is a possible correlation between these probabilities; for
example, treatment discontinuation because of intolerable side effects is likely to be
related to the risk for acute EPS. Such potential correlation between these parameters
has not been considered in the analysis. On the other hand, the correlations across
probability of relapse, probability of treatment discontinuation because of intolerable
side effects and probability of treatment discontinuation because of other reasons
have been taken fully into account because data on these three parameters were
analysed together in a competing risks mixed treatment comparison model. The
posterior simulations resulting from this exercise were then exported jointly and
fitted into the Excel file of the economic model where the probabilistic analysis was
implemented.
The analysis adopted the perspective of the NHS and personal social services, as
recommended by NICE. Costs associated with the pharmacological treatment of
people with schizophrenia were estimated by combining data from the NHS and
other national sources of healthcare resource utilisation, as well as information from
published studies conducted in the UK, with national unit costs. A number of
further GDG estimates and assumptions were required to inform the cost parameters
of the economic model. The results of the economic analysis demonstrated that drug
acquisition costs do not determine the relative cost effectiveness of antipsychotic
medications: haloperidol had the lowest probability of being cost effective in
probabilistic analysis, despite the fact that it is by far the cheapest drug among those
assessed. On the other hand, paliperidone was ranked highly in terms of cost
effectiveness (the third best option in deterministic analysis at 10 years and over a
lifetime; and the second highest probability of being cost effective in probabilistic
analysis), despite having the highest acquisition cost. Although drug acquisition
costs seem to be unimportant in determining cost effectiveness, it must be noted that
the prices of a number of antipsychotic medications are expected to fall in the future
because more drugs will be available in generic form.
Deterministic analysis showed that the probability of relapse was the key driver of
cost effectiveness. It is not surprising, therefore, that zotepine, which was shown to
be the most cost-effective option in both deterministic and probabilistic analyses,
had the lowest average probability of relapse and the highest probability of being the
most effective drug in reducing relapse in the mixed treatment comparison analysis;
olanzapine and paliperidone, which were the second and third most cost-effective
options in deterministic analysis, respectively, had the third and second lowest
relapse rates, respectively, and were ranked third and second best drugs in reducing
relapse, respectively (details of effectiveness ranking in mixed treatment comparison
analysis are provided in Table 115). These findings indicate that it is the effectiveness
of an antipsychotic drug in preventing relapse that primarily affects its cost
effectiveness, especially considering that the rates of side effects were not shown to
have any significant impact on the cost-effectiveness results; such a hypothesis
seems reasonable, given that relapse prevention greatly improves the HRQoL of
people with schizophrenia and, simultaneously, leads to a substantial reduction in
hospitalisation rates and associated high costs. In fact, reduction in inpatient costs
associated with the development of acute episodes affects the level of total costs
associated with antipsychotic medication and the ranking of options in terms of cost
effectiveness in the long term, as shown in sensitivity analysis.
Besides the health and social care costs that were considered in this analysis,
according to the NICE recommended economic perspective, wider societal costs
(such as costs borne to the criminal justice system, personal expenses of people with
schizophrenia and their carers, productivity losses of people with schizophrenia,
carers’ time spent with people with schizophrenia, which may also translate to
productivity losses for carers, as well as the emotional burden associated with
schizophrenia) need to be taken into account when the cost effectiveness of
antipsychotic medications is assessed.
11.5 CONCLUSIONS
The economic analysis undertaken for this guideline showed that zotepine may be
potentially the most cost-effective antipsychotic medication among those assessed
for relapse prevention in people with schizophrenia in remission. However, results
were characterised by high uncertainty, and probabilistic analysis showed that no
antipsychotic medication can be considered to be clearly cost effective compared
with the other options included in the assessment: the probability of each
intervention being cost effective ranged from roughly 5% (haloperidol) to about 27 to
30% (zotepine), and was independent of the cost-effectiveness threshold used and
the time horizon of the analysis (that is, 10 years or a lifetime). The probability of 27
to 30% assigned to zotepine, although indicative, is rather low and inadequate to
lead to a safe conclusion regarding zotepine’s superiority over the other
antipsychotic medications assessed in terms of cost effectiveness. In addition, clinical
data for zotepine in the area of relapse prevention (as well as for paliperidone and
aripiprazole) came from a single placebo-controlled trial. Data on side effects were
not comprehensive; in particular, data on the risk for diabetes and glucose
intolerance associated with use of antipsychotic medications were sparse, so that the
impact of the risk for diabetes and its complications on the relative cost effectiveness
of antipsychotic drugs could not be determined accurately. It has to be noted,
however, that the estimated rates of side effects considered in the analysis did not
significantly affect the cost effectiveness results.
Further research is needed on the benefits and patterns of use of antipsychotic

medications in the area of relapse prevention in people with schizophrenia that is in
remission, as well as on the rates of associated long-term metabolic side effects, to
address the uncertainty characterising the results of the economic analysis.
Moreover, clinical data in the area of relapse prevention are needed for quetiapine
and FGAs other than haloperidol, to enable a more comprehensive assessment of the
relative cost effectiveness of antipsychotic medications in relapse prevention for
people with schizophrenia that is in remission.**2009**
12 TEAMS AND SERVICE-LEVEL
INTERVENTIONS
12.1 INTRODUCTION
This chapter fully updates the review of teams and service-level interventions
(developed as part of ‘community care’ in different parts of the world, as well as
those specifically developed in the UK) from the 2002 and 2009 guidelines. The GDG
recognised that much of the research in this area has followed changes in practice,
often led by policy initiatives to move from hospital to community care, with mental
health service providers developing different, previously untested, service
configurations in the community as an alternative to relatively costly inpatient
settings.
Some teams and services have been developed for the routine, non-acute provision
of care for people with psychosis and schizophrenia in community settings, for
example, community mental health teams (CMHTs), while others have focused
much more on treatment during times of crisis that, previously, would have led to
an inpatient admission, for example, crisis resolution and home treatment teams
(CRHTTs). The latter have, in the main, been designed as alternatives to acute
hospital care. Some services have, nevertheless, been designed to both support
people day to day in the community, and provide some treatment and care either to
prevent an impending crisis or even to avoid acute admission, for example, assertive
community treatment (ACT). To reduce confusion and in the service of clarity, the
GDG has synthesised the available evidence to provide guidance about the best team
and service-level interventions for acute and non-acute care in community settings.
The GDG, therefore, considered and reviewed the evidence for non-acute
community-based care and the evidence for acute or crisis community-based care
separately. Although the provision of non-acute and acute/crisis care is not always
clearly demarcated within mental health and social care services in practice, the
trials contributing to these two reviews were nevertheless separated. The GDG also
considered the importance of reducing the duration of untreated psychosis (DUP)
for people with first episode psychosis because longer DUP has been reported to be
associated with poorer outcomes (Marshall et al., 2005; Perkins et al., 2005), and
much of the rationale for the emergence of early intervention services (EIS; also
known as ‘early intervention in psychosis services’) was based on reducing DUP.
The GDG utilised the review by Lloyd-Evans et al. (2011) to assess the effectiveness
of programmes that aim to reduce DUP.
The chapter is thus divided into three sections. Section 12.2 discusses the interface
between primary and secondary care in relation to service provision. Section 12.3
reviews non-acute community mental healthcare and includes an evaluation of EIS

and early detection programmes to reduce DUP, CMHTs and intensive case
management (ICM—a recent term that encompasses ACT and case management).
Section 12.4 reviews community-based alternatives to acute admission and includes
CRHTTs, crisis houses and acute day hospital care.
In reviewing the evidence for the effectiveness of different services in the 2002
guideline, the GDG decided to focus on the RCT because this is the best design to
evaluate the effectiveness of competing interventions. However, team and service-
level interventions are essentially complex interventions including, for example,
psychological interventions combined with specific team operating protocols and
case load limits. The GDG has ensured that wherever meta-analyses have been
performed, the definition of the team or service-level intervention has been
examined carefully. Moreover, it is important to recognise that it is often difficult to
establish with certainty, in a simple RCT, what aspects of the team or service-level
intervention are the effective ingredients. In this regard, the GDG has played an
important consensus-based role in grouping different types of intervention to allow
meta-analysis and in interpreting the findings for each set of comparisons.
Individual randomisation is not possible in studies of early detection programmes,

which by definition, target whole populations from which people with first episode
psychosis might be referred to services. Therefore, the review of interventions to
reduce DUP was not limited to RCTs.
Many of the studies have been undertaken outside the UK. Where the comparator is
standard care, the GDG has taken this into consideration because ‘standard care’ is
often different in important respects in other countries. Where UK studies have been
available, the GDG has looked at UK sub-analyses alongside the full dataset analysis.
The GDG also considered the 2002 and 2009 guidelines in the area of primary care
and the interface between primary and secondary care, both areas being the subject
of a number of consensus-based recommendations. The GDG for the 2014 guideline
has added to these recommendations, mainly in the area of physical health, and has
also retained and modified some of the considerations made by the GDGs for the
2002 and 2009 guidelines, both within the text and the associated recommendations.
12.2 INTERFACE BETWEEN PRIMARY AND SECONDARY

CARE
12.2.1 Introduction
This section focuses on the initial pathway to specialist help for a person presenting
with first episode psychosis to primary care; and those with an established diagnosis
managed either collaboratively between primary and secondary care, or wholly in
primary care. The recommendations are based on an updated consensus-based
narrative synthesis of the relevant sections of Psychosis and Schizophrenia in Children
and Young People (NCCMH, 2013 [full guideline]; NICE, 2013a) and the 2009 adult
guideline (NICE, 2009d).
12.2.2 First episode psychosis and its presentation
The emerging distress of a first episode of psychosis will cause many people, often
supported by their families, to seek help from their GP. However, this is an
infrequent event for an individual GP, who on average encounters around one to
two patients per year with a suspected emerging psychosis (Simon et al., 2005);
frequency is slightly increased in inner city areas. Notwithstanding this low
frequency, GPs are the most common referral agents to specialist services, and,
furthermore, their involvement is also associated with reduced use of the Mental
Health Act (Burnett et al., 1999) making their role important in detecting psychosis
and initiating the pathway to specialist care.
Psychosis is difficult for GPs to recognise, and there a number of reasons for this. It
tends to occur for the first time when people are young: more than three quarters of
men and two thirds of women who experience psychosis have their first episode
before the age of 35. Indeed, most first episodes occur between late teens to late
twenties, mirroring when many other lifetime mental disorders present for the first
time (Kessler et al., 2007) and against a backdrop of increasing psychological distress
for many young people -- for instance, 20% of young people will experience a
diagnosable depressive episode by the age of 18 years (Lewinsohn et al., 1993).
Moreover, serious disorders like psychosis often start off like milder and far more
common mental health problems, and rarely present initially with clear cut
psychotic symptoms. The challenge, therefore, for GPs in detecting psychosis
promptly is to distinguish its presentation at an early undifferentiated phase and at
an age when many people may first present with psychological difficulties. When
asked how to improve detection of emerging first episode psychosis, GPs request
better collaboration with specialist services and low-threshold referral services rather
than educational programmes (Simon et al., 2005).
In view of the evidence presented in this guideline regarding suspected psychosis

(that early treatment with CBT may decrease the likelihood of transition to psychosis
whereas antipsychotics appear to be ineffective) and first episode psychosis (that
there are benefits for being seen at an early stage), the GDG regarded the role of the
GP in recognising and monitoring both suspected and likely symptoms of psychosis
to be a clear focus for developing consensus-based recommendations.
The GDG therefore concluded that people presenting with symptoms of suspected
or actual psychosis in primary care should be referred to EIS.
After the first episode, some people refuse to accept the diagnosis and sometimes
also reject the treatment offered. Bearing in mind the consequences of a diagnosis of
psychosis and schizophrenia, many people in this position, perhaps unsurprisingly,
want a second opinion from another consultant psychiatrist. This is often requested
through a person’s GP if a person knows it is available.

12.2.3 People with an established diagnosis of psychosis and
schizophrenia in primary care
The GDG for the 2009 guideline made the following statements, which underpin a
number of recommendations about primary care (the GDG for the 2014 guideline
decided to only modify the related recommendations to improve the wording and to
extend physical healthcare; see section below on physical health):
‘People with an established diagnosis of schizophrenia who are managed in primary

care require regular assessment of their health and social needs. This should include
monitoring of mental state, medication use and adherence, side effects, social
isolation, access to services and occupational status. All such people should have a
care plan developed jointly between primary care and secondary mental health
services. Regular monitoring of physical health is also essential. With consent from
service users, non-professional carers should also be seen at regular intervals for
assessment of their health and social care needs. Carers should also be offered an
assessment of their needs.
Advance statements and advance decisions about treatment should be documented

in the service user’s notes. These should be copied from secondary services to the
responsible GP. If no secondary service is involved in the service user’s care (because
they have recently moved to the area, for example), the GP should ensure that any
existing advance decisions or statements are copied to the secondary services to
whom referral is made.
When a person with schizophrenia is planning on moving to the catchment area of a

different NHS trust, their current secondary care provider should contact the new
secondary and primary care providers, and send them the current care plan.
People presenting to primary care services who are new to the area (not known to
local services) with previously diagnosed psychosis should be referred to secondary
care mental health services for assessment, subject to their agreement. The GP
should attempt to establish details of any previous treatment and pass on any
relevant information about this to the CMHT.
When a person with schizophrenia is no longer being cared for in secondary care,
the primary care clinician should consider re-referral of the service user to secondary
care. When referring a service user to secondary mental health services, primary care
professionals should take the following into account:
• Previous history: if a person has previously responded effectively to a

particular treatment without experiencing unwanted side effects and is
considered safe to manage in primary care, referral may not be necessary.
• Views about referral: the views of the mental health service user should be
fully taken into account before making a referral. If the service user wants to
be managed in primary care, it is often necessary to work with the family and
carers. Sharing confidential information about the service user with carers

raises many ethical issues, which should be dealt with through full discussion
with the service user.
• Non-adherence to treatment: this may be the cause of the relapse, possibly as
a result of lack of concordance between the views of the service user and of
the healthcare professionals, with the former not recognising the need for
medication. Alternatively, non-adherence might be the consequence of side
effects. Finding the right antipsychotic drug specifically suited to the service
user is an important aim in the effective management of schizophrenia.
• Side effects of medication and poor response to treatment: the side effects of
antipsychotic drugs are personally and socially disabling, and must be
routinely monitored. Side effects are also a cause of poor response to
treatment. For about 40% of people given antipsychotics, their symptoms do
not respond effectively.
• Concerns about comorbid drug and alcohol misuse: substance misuse by
people with schizophrenia is increasingly recognised as a major problem,
both in terms of its prevalence and its clinical and social effects (Banerjee et
al., 2002). Monitoring drug and alcohol use is an essential aspect of the
management of people with schizophrenia in primary and secondary care.
• Level of risk to self and others: people with schizophrenia, especially when
relapse is impending or apparent, are at risk of suicide and are often
vulnerable to exploitation or abuse. During an acute episode of illness,
conflicts and difficulties may manifest themselves through social disturbances
or even violence.’
The GDG for the 2014 guideline wished to add the following bullet point to this list:
• General social functioning and self-care: loss of employment/vocational

activity, social withdrawal, self-neglect, and financial or housing difficulties
can all be signs of or precursors to relapse. Social exclusion is a common
feature in people with psychosis or schizophrenia diagnosis. Referral to
secondary mental health services or other relevant agencies may be required.
The 2009 guideline concluded by saying: ‘The identification of patients with

schizophrenia in a well-organised computerised practice is feasible (Kendrick et al.,
1991; Nazareth et al., 1993). The organisation and development of practice case
registers is to be encouraged because it is often the first step in monitoring people
with schizophrenia in general practice. There is evidence that providing payment
incentives to GPs leads to improved monitoring of people with schizophrenia (Burns
& Cohen, 1998). In 2004, as a part of the GP contract, the Quality and Outcomes
Framework was introduced in English general practice as a voluntary process for all
general practices – schizophrenia is one of the medical conditions to be monitored as
part of this framework’ (NCCMH, 2010 [full guideline]).
Physical health
Since the 2009 guideline, the evidence base for physical ill health among people with
psychosis and schizophrenia has continued to develop. In particular, more

understanding of why cardiovascular disease occurs at such high rates in people
with schizophrenia makes it appropriate to review the existing recommendations
relating to physical healthcare in primary care. New recommendations about
lifestyle interventions to reduce the impact of cardiovascular risks are described in
Chapter 10. In considering such interventions it is also necessary to reflect on the
adequacy of screening for cardiovascular risk factors and, related to this, monitoring
for adverse cardiometabolic effects from antipsychotic medication.
People with psychosis and schizophrenia are at considerably increased risk of poor
physical health. Although suicide accounts for a quarter of all premature mortality
in people with severe mental ill health, including schizophrenia, of all causes of
premature death, cardiovascular disease is now the commonest in this group. This
tendency is no doubt a result of a complex combination of social exclusion, poor
diets, high rates of obesity, lack of physical activity and high rates of smoking,
compounded by health risks linked to genetic vulnerabilities and adverse effects of
antipsychotic medication. These various factors lead to more frequent disturbances
of glucose and lipid metabolism, resulting in atherosclerosis. The rate of diabetes
mellitus is two to three times higher than for the general population (almost entirely
accounted for by type 2 diabetes). A European study screening people with
schizophrenia who were not known to have diabetes, discovered 10% had type 2
diabetes and 38% were at high risk of type 2 diabetes; this population’s average age
was only 38 years (Manu et al., 2012).
Concerns about cardiovascular mortality more generally have attracted a public

health focus in the UK over the last 2 decades. For instance, health promotion and
disease management programmes for conditions like heart disease and diabetes
have become established in primary care, further encouraged since 2006 through the
primary care pay for performance scheme, the Quality and Outcomes Framework
(NHS Employers, 2011). Although there have been reductions in cardiovascular
morbidity and mortality in the general population, these benefits have not been
enjoyed by people with severe mental illness—indeed the mortality gap between the
general population and people with severe mental illness may still be widening
(Brown et al., 2010). It is important to recognise, then, that some of the key
antecedent risks for premature mortality in this group may emerge and become
established early in the course of psychosis, perhaps even in or before the first
episode.
People with first episode psychosis, exposed for the first time to antipsychotics, are
particularly vulnerable to rapid weight gain (Alvarez-Jimenez et al., 2008; Kahn et
al., 2008) and adverse cardiometabolic disturbance (Foley & Morley, 2011). The
subsequent trajectory of weight gain and increasing metabolic disturbance, when
combined with high rates of tobacco smoking even before the first episode begins
(Myles et al., 2012), provide a potent mix of cardiovascular risk factors. Given that
modifiable cardiovascular risk appears within months of commencing treatment
(Foley & Morley, 2011), the onus should arguably shift towards a prevention and

early intervention approach to cardiovascular risk (Phutane et al., 2011). The GDG
accepted this view.
A prerequisite for successful prevention is the implementation of guidelines such as

the European screening and monitoring guidelines for diabetes and cardiovascular
risk in schizophrenia (De Hert et al., 2009a). Yet despite numerous published
screening recommendations, monitoring rates remain poor in adults (Buckley et al.,
2005; Mackin et al., 2007b; Morrato et al., 2009; Nasrallah et al., 2006). This was
recently also confirmed in the UK by the National Audit of Schizophrenia (Royal
College of Psychiatrists, 2012). Importantly, this audit examined the implementation
of recommendations for physical health monitoring set out in the 2009 guideline for
people under the care of mental health services in community settings during the
previous 12 months. Ninety-four per cent of mental health trusts across England and
Wales participated in an audit of over 5,000 patients’ case records making it very
likely that its findings reflect current practice. On average, only 28% of this
population (range by mental health trust of 13 to 69%) had a recorded assessment of
the main risk factors for cardiovascular disease (BMI, smoking status and blood
pressure, glucose and lipids) within the previous 12 months. The findings of the
audit suggest inconsistent and often inadequate local monitoring arrangements and
indicate a need to establish greater clarity over responsibilities and improve
communication between primary and secondary care.

The GDG for the 2014 guideline reconsidered the 2002 and 2009 guidelines in the
area of primary care and the primary and secondary care interface. It was agreed
that although there is no robust evidence to guide recommendations in this area, the
GDG for the 2014 guideline concurred with its predecessors that consensus-based
recommendations (based on the considerations above but not restricted to them)
should be developed to help guide primary and secondary care health and social
care professionals in these areas. Service users with serious mental illness tend to be
forgotten in primary care, by both primary and secondary care professionals, and
there is a relatively low level of understanding of the role of primary care in the
initial management of psychosis and schizophrenia, for example, when and if
antipsychotic medication should be introduced. Moreover, the breadth and depth of
initial assessments of people with psychosis and schizophrenia on entry to
secondary care are very variable, as are the development and role of care plans.
Service users commonly do not know that they have a care plan, especially when
they first use secondary care services. Many service users like to return to primary
care when they are stable, and primary care professionals are often unsure about
their role in this context, nor about when to reengage secondary care and to re-refer.
Finally, when service users move house, this often involves changing both primary
and secondary care services. Service users frequently become lost to services at this
point. The GDG for the 2014 guideline decided to follow the GDG for the 2009
guideline and include a recommendation about how to minimise loss from services
at this point. Advance warning and relevant information from existing care
providers should be given to the new providers.

It should be recognised that, of all parts of the care pathway for people with
psychosis and schizophrenia, the role of primary care and the management of the
primary-secondary care interface are areas of weakness and are relatively
inaccessible to robust research. Primary care and its interface with secondary care
are both important and yet lacking in evidence for best practice. In addition, there is
no health economic evidence in these areas. As such, the following
recommendations are intended to minimise harm, improve assessment, prevent
service users becoming lost to services and ensure that when problems arise in
primary care service users can gain access easily to the services they need.
At present, for most GPs, between one and two of the people on their list each year
will develop a first episode psychosis. In these circumstances, referral to EIS appears
to produce most benefit for the service user (for the review of EIS see Section 12.3.2).
However, some GPs, on seeing a person with a psychotic presentation, consider the
use of antipsychotics as a first step, while others are uncertain. In some situations,
this may well be the right intervention, especially if the service user is very
distressed or the psychosis is well advanced. However, given the increasing
availability and preference for psychological treatments, the sometimes severe side
effects that can occur with first exposure to antipsychotics, and the preparatory
investigations that are usually necessary before starting these drugs, the GDG
decided to recommend that antipsychotics should not be started in primary care
without prior discussion with a consultant psychiatrist.
A further area of variable practice includes the assessment of service users on arrival
in secondary care. Entering secondary care for the first time is a very important
experience for service users and can colour future attitudes to secondary care.
Professionals usually take this into account. However, this can lead to assessments
being relatively brief and/or limited in content. It is also important to bear in mind
that some drugs can precipitate a psychosis and that psychoses are often associated
with coexisting physical and mental health problems. The GDG decided to
adumbrate the key areas that should be covered in the assessment, so as to ensure
that, even if these areas cannot be covered immediately, professionals in secondary
care should aim for a genuinely comprehensive assessment over time. After all,
psychosis and schizophrenia affects the whole of a person’s life, including
relationships, physical activity and health, education and employment, and their
ability to pursue individual goals; and even where symptoms may be less severe, it
is important to get a baseline of personal functioning at the point of admission to
secondary care so as to track changes that may well come about through the acute
episode and after recovery.
With these considerations in mind, the GDG recommended that the assessment in
secondary care should include a full psychiatric assessment, as well as a full medical
assessment for physical ill health and the possibility of organic factors influencing
the development of the psychosis. Physical assessment should include smoking
status, nutrition, physical activity and sexual health, all of which are commonly

affected either early on (for example 59% of people with a first episode of psychosis
are already smoking) or certainly later (people with established schizophrenia have
high rates of cardiovascular disease). People with psychosis and schizophrenia will
experience considerable disruption to their social and psychological life. Assessment
should include looking at their accommodation, their capacity to engage in cultural
activities appropriate to their ethnicity, and to understand the burdens they have in
terms of caring for others, including children or parents. It should also include
evaluation of their social networks, relationships and possible personal trauma, and
also neurodevelopmental considerations, especially for younger users of EIS who
have an increased risk of presenting with social, cognitive and motor impairments.
Psychosis will affect a person’s quality of life, activities of daily living and access to
employment, all of which need to be included in the assessment. It is common for
people with psychosis to experience quite marked anxiety, depression and alcohol or
drug (both street bought and prescribed) misuse; comorbidities can occur at any
time but especially early on in the psychosis. Engaging service users is also a
particular problem, especially in the early period. The GDG considered it helpful to
make the assessment and development of a written care plan a focus for engagement
by undertaking this jointly with the service user, wherever this is possible. The care
plan should include all the issues identified in the assessment.
When a person presents for the first time, or even over the first few times, it may be
quite clear that they have developed a psychosis, but not so clear whether they have
schizophrenia, bipolar disorder or other affective psychosis, or another less common
form of psychosis. This diagnostic problem is made all the more difficult by the
coexistence of other mental health problems. Nevertheless, it usually becomes
apparent that the psychosis is either a schizophrenic psychosis or an affective
psychosis, and the relevant guidelines should be followed for the latter, whether this
is the Bipolar Disorder (NICE, 2006a) or Depression (NICE, 2009a) guideline.
Most psychotic episodes resolve within 6 to 8 months, although it can take

substantially longer for some people to reach stability. After a psychosis has resolved
and the person is stable, it is common that service users wish to be discharged back
to primary care. This transfer should be supported by secondary health and social
care professionals who need to contact primary care and arrange transfer of care
plans, if this has not occurred already. Primary healthcare professionals should
ensure that, when a person first returns from secondary care services to primary
care, they are added to a case register of all people with psychosis within their
practice. This is a key step in ensuring that people with psychoses receive the right
mental and physical healthcare within primary care.
It is important to recognise that antipsychotics can have quite severe and unpleasant
side effects which, if carefully managed, can be minimised or even prevented. If they
become excessive or intolerable, this can lead to service users stopping treatment
altogether, sometimes suddenly, provoking relapse. It is, therefore, important to
monitor side effects in primary care. It is also important to monitor psychotic
symptoms in primary care, and to keep an eye on common accompaniments to

possible relapse such as an increase in alcohol consumption or drug taking. If there
is concern in primary care, the care plan should be consulted. The care plan should
include a crisis plan and the name of either the key clinician (which may be a
consultant psychiatrist or psychologist or other secondary health or social care
professional) and/or the care coordinator. Primary care professionals should not
hesitate in making direct contact for advice and in making a referral. Key factors that
should encourage referral include any factor associated with an increased likelihood
of relapse, such as persisting psychotic symptoms (a poor response to treatment), a
failure to continue with agreed treatment, intolerable or very unpleasant side effects,
substance misuse and a risk of self-harm or harm to others. However, some service
users and/or their carers will request re-referral to secondary care, usually because
they want their drug regime reviewed because of side effects, such as excessive
drowsiness or sexual side effects, or for specialist psychological treatments for
psychosis. Requests for re-referral should be enabled and supported.
A comprehensive multidisciplinary assessment and close monitoring of people with

psychotic symptoms would ensure timely detection and appropriate management of
physical ill health. There is no health economic evidence in this area for people with
psychosis and schizophrenia; however the GDG felt that since psychosis and
schizophrenia affect the whole of a person’s life, and people with these conditions
are at considerable increased risk of poor physical health, that preventing ill health
(including cardiovascular disease) and premature death, and minimising the adverse
effects associated with antipsychotic medication, have clear potential to reduce
healthcare costs and lead to improvements in health related quality of life.

12.2.5.1 Do not start antipsychotic medication for a first presentation of sustained
psychotic symptoms in primary care unless it is done in consultation with a
consultant psychiatrist. [2009; amended 2014]
12.2.5.2 Carry out a comprehensive multidisciplinary assessment of people with
psychotic symptoms in secondary care. This should include assessment by a
psychiatrist, a psychologist or a professional with expertise in the
psychological treatment of people with psychosis or schizophrenia. The
assessment should address the following domains:
• psychiatric (mental health problems, risk of harm to self or others, alcohol
consumption and prescribed and non-prescribed drug history)
• medical, including medical history and full physical examination to identify
physical illness (including organic brain disorders) and prescribed drug
treatments that may result in psychosis
• physical health and wellbeing (including weight, smoking, nutrition, physical
activity and sexual health)
• psychological and psychosocial, including social networks, relationships and
history of trauma
• developmental (social, cognitive and motor development and skills, including
coexisting neurodevelopmental conditions)

• social (accommodation, culture and ethnicity, leisure activities and recreation,
and responsibilities for children or as a carer)
• occupational and educational (attendance at college, educational attainment,
employment and activities of daily living)
• quality of life
• economic status. [2009; amended 2014]
12.2.5.3 Routinely monitor for other coexisting conditions, including depression,

anxiety and substance misuse particularly in the early phases of treatment.
[2009; amended 2014]
12.2.5.4 Write a care plan in collaboration with the service user as soon as possible
following assessment, based on a psychiatric and psychological formulation,
and a full assessment of their physical health. Send a copy of the care plan to
the primary healthcare professional who made the referral and the service
user. [2009; amended 2014]
12.2.5.5 If the person’s symptoms and behaviour suggest an affective psychosis or
disorder, including bipolar disorder and unipolar psychotic depression,
follow the recommendations in Bipolar disorder (NICE clinical guideline 38)
or Depression (NICE clinical guideline 90). [new 2014]
12.2.5.6 Offer people with psychosis or schizophrenia whose symptoms have
responded effectively to treatment and remain stable the option to return to
primary care for further management. If a service user wishes to do this,
record this in their notes and coordinate transfer of responsibilities through
the care programme approach. [2009]
12.2.5.7 Develop and use practice case registers to monitor the physical and mental
health of people with psychosis or schizophrenia in primary care. [2009]
12.2.5.8 When a person with an established diagnosis of psychosis and
schizophrenia presents with a suspected relapse (for example, with
increased psychotic symptoms or a significant increase in the use of alcohol
or other substances), primary healthcare professionals should refer to the
crisis section of the care plan. Consider referral to the key clinician or care
coordinator identified in the crisis plan. [2009]
12.2.5.9 For a person with psychosis or schizophrenia being cared for in primary
care, consider referral to secondary care again if there is:
• poor response to treatment
• non-adherence to medication
• intolerable side effects from medication
• comorbid substance misuse
• risk to self or others. [2009]
12.2.5.10 When re-referring people with psychosis or schizophrenia to mental
health services, take account of service user and carer requests, especially
for:
• review of the side effects of existing treatments
• psychological treatments or other interventions. [2009]
12.2.5.11 When a person with psychosis or schizophrenia is planning to move to
the catchment area of a different NHS trust, a meeting should be arranged
between the services involved and the service user to agree a transition plan
before transfer. The person’s current care plan should be sent to the new
secondary care and primary care providers. [2009]
12.3 NON-ACUTE COMMUNITY MENTAL HEALTHCARE

12.3.1 Introduction
After the decline of the asylum and before the development of modern day
community services, many mental health services provided a fairly typical medical
arrangement based upon hospital care and outpatient clinics, with some facility for
day care for people with a chronic illness and/or severe impairment. Prior to the
development of community care, non-acute (routine, scheduled or planned) care
took place predominantly in outpatient clinics or day services, and sometimes in
hospital, in specific situations, for example, when medication changes in a well
patient had the potential to destabilise their condition.
However, following an acute episode of psychiatric illness, discharging patients

often proved problematic as there were little or no facilities to provide more
supportive community-based help closer to people’s homes. To enhance discharge,
community psychiatric nurse roles, based on psychiatric wards and helping people
settle in the community, were developed in the 1960s to provide an intermediate
level of support away from hospital. By the mid 1990s community-based teams
emerged to provide more routine care and to help avoid acute care when higher
levels of support and treatment were needed. Although CMHTs became the routine,
with consultant psychiatrists bridging the gap between non-acute community care
and more clearly acute hospital care, there was surprisingly little evidence to suggest
that CMHTs were any better or any worse than the previous arrangement of
services. Nevertheless, service users generally prefer non-hospital-based solutions if
they are given the choice.
With pressure on resources and national policy to move away from big hospitals,
and a more explicit acceptance that service users wanted to access services for
routine care in the community, new teams/services were formed, such as acute day
hospitals, ACT, case management and ICM and later, EIS for people with early
psychosis (for the first 3 years). This section of the guideline reviews the evidence for
the clinical and cost effectiveness of EIS, CMHTs and ICM as providers of
(predominantly) non-acute care, and also early detection programmes to reduce
DUP. It should be remembered, however, that EIS will often accept patients with
early schizophrenia in a crisis, usually with support from other acute, community-
based services; and ICM often provides crisis care for some of their service users.

12.3.2 Early intervention services
Introduction
The NHS Plan (Department of Health, 2000) set out a requirement for mental health
services to establish EIS. EIS are expected to provide care for: (a) people aged
between 14 and 35 years with a first presentation of psychotic symptoms; and (b)
people aged 14 to 35 years during the first 3 years of psychotic illness. The Mental
Health Policy Implementation Guide (Department of Health, 2001) set out a wide range
of tasks for EIS, including: reducing stigma and raising awareness of symptoms of
psychosis; reducing DUP; promoting better engagement with treatment and
services; providing evidence-based treatments; promoting recovery for young
people who have experienced an episode of psychosis; and working across the
traditional divide between CAMHS and AMHS, as well as in partnership with
primary care, education, occupational therapy, social services, youth and other
services. EIS was an innovation introduced over the last 10 to 15 years as a
progressive, integrating service able to provide a broad range of effective treatments
with the explicit aim of better engaging young people with psychosis, reducing time
to treatment and minimising impairment. However, at the time of their national
introduction, there was no RCT evidence for their effectiveness compared with
standard care, either in the UK or elsewhere.
Early intervention is primarily concerned with identification and initial treatment of

people with psychotic illnesses, such as schizophrenia. Identification may be
directed either at people in the prodromal phase of the illness (‘earlier early
intervention’, or prevention) or at those who have already developed psychosis
(‘early intervention’). Early identification of people with psychotic disorders may be
especially relevant to specific groups, for example, African–Caribbean people who
are at higher risk of developing a psychosis and presenting very late in the course of
the illness. Central to the rationale for early identification is the concept of DUP. The
sooner the psychosis is identified the sooner the psychosis can be treated. A number
of researchers have reported that the longer the psychosis goes untreated, the poorer
the prognosis becomes (Loebel et al., 1992; McGorry et al., 1996). This finding has led
them to argue that new services are required to reduce the length of time that people
with psychosis remain undiagnosed and untreated. The GDG therefore decided to
examine the evidence for EIS or any other intervention, including public awareness
campaigns and GP awareness and education programmes, to improve detection of
psychosis with consequent reduction in DUP (see Section 12.3.3).
Definition and aim of intervention/ service system

EIS is defined as a service approach with focus on the care and treatment of people
in the early phase (usually up to 5 years) of psychosis or schizophrenia, sometimes
including the prodromal phase of the disorder. The service may be provided by a
team or a specialised element of a team, which has designated responsibility for at
least two of the following functions:

• early identification and therapeutic engagement of people experiencing a first
episode of psychosis
• provision of age-appropriate, evidence-based pharmacological and
psychosocial interventions during and following a first episode psychosis
• education of the wider community to reduce obstacles to early engagement in
treatment.
Clinical review protocol (early intervention services)

using meta-analysis, and where data were lacking, the available evidence was
synthesised using narrative methods.

Table 132: Clinical review protocol summary for the review of early intervention
services
potential harms of early intervention services compared with treatment as
usual or another intervention?
Objectives To evaluate the clinical effectiveness of early intervention services in the
Intervention(s) Early intervention services
Critical outcomes • Adverse events
o Suicide
• Functioning disability
• Service use
o Hospitalisation (admissions, days)
o In contact with services
• Response /relapse
o Total symptoms
o Positive symptoms
o Negative symptoms
• Employment and education
o Competitive employment
o Occupation (any)
o Attendance at school/college
• Duration of untreated psychosis
• Carer satisfaction
Process
Date searched SR/ RCT: 2002 to June 2013
Study design RCT
Sub-analysis

studies.

Studies considered 41
Four RCTs (N = 800) met the eligibility criteria for this review: CRAIG2004B (Craig
et al., 2004), GRAWE2006 (Grawe et al., 2006), KUIPERS2004 (Kuipers et al., 2004)
and PETERSEN2005 (Petersen et al., 2005). All were published in peer-reviewed
journals between 2004 and 2006 and were conducted in the UK or Europe. Further
15a.
All four eligible trials included sufficient data to be included in statistical analysis
and compared EIS with standard care. The proportion of individual with psychosis
and schizophrenia ranged from 93 to 100%. The length of treatment ranged from 52
to 104 weeks and only two trials had medium-term follow-up data. Table 133
provides an overview of the included trials.
Table 133: Study information table for trials included in the meta-analysis of EIS
versus any alternative management strategy
Early intervention services versus any alternative

management strategy
Total no. of trials (k); participants (N) k = 4; N = 800
Study ID(s) CRAIG2004B
GRAWE2006
KUIPERS2004
PETERSEN2005
Country Denmark (k = 1)
Norway (k = 1)
UK (k = 2)
Year of publication 2004- 2006
Mean age of participants (range) 26.5 years (25.4 to 27.8 years)
Mean percentage of participants with primary 98.31% (93.22 to 100%)
diagnosis of psychosis or schizophrenia (range)
Mean percentage of women (range) 34.52% (23.73 to 40.95%)
Length of follow-up (range) 52 to 104 weeks
Intervention type Croydon Outreach and Assertive Support Team (k = 1)
Integrated Treatment (k = 2)
Specialised care group - assertive outreach for early
psychosis (k = 1)
Comparisons Standard treatment (k = 4)
Clinical evidence for the review of early intervention services verses any
control

Table 134: Summary of findings table for EIS versus any alternative management
strategy

Intervention: EIS
Control EIS CI) (GRADE)
Adverse events 14 per 4 per 1000 RR 0.27 691 ⊕⊕⊕⊝
(suicide, actual 1000 (1 to 24) (0.05 to (2 studies) Moderate1
and attempted) - 1.65)
end of treatment
Adverse events 15 per 11 per 1000 RR 0.74 547 ⊕⊕⊕⊝
(suicide, actual 1000 (2 to 48) (0.17 to (1 study) Moderate1
and attempted) - 3.28)
>12 months’
follow-up
Service use 674 per 593 per 1000 RR 0.88 733 ⊕⊕⊕⊝
(hospitalisation) - 1000 (533 to 661) (0.79 to (3 studies) Moderate1
end of treatment 0.98)
Service use N/A Mean service use (hospitalisation, number of N/A 683 ⊕⊕⊕⊝
(hospitalisation, bed days - end of treatment) in the intervention (2 studies) Moderate1
number of bed groups was 0.18 standard deviations lower
days) - end of (0.33 to 0.03 lower)
treatment
(hospitalisation, admissions - end of treatment) in the (1 study) Moderate1
number of intervention groups was 0.46 standard
admissions) - end deviations lower (0.8 to 0.12 lower)
of treatment
(hospitalisation) - 1000 (348 to 495) (0.78 to (2 studies) Moderate1
>12 months’ 1.11)
follow-up
(hospitalisation, bed days, >12 months’s follow-up) in the (2 studies) Moderate1
number of bed intervention groups was 0.08 standard
days) - >12 deviations lower (0.24 lower to 0.07 higher)
months’ follow-up
(hospitalisation, admissions, >12 months’ follow-up) in the (1 study) Moderate1
number of intervention groups was 0.2 standard
admissions) - >12 deviations lower (0.6 lower to 0.2 higher)
months’ follow-up
(contact - not in 1000 (63 to 147) (0.4 to (2 studies) Moderate1
contact with 0.93)
index team) - end
of treatment

(contact - not in 1000 (85 to 288) (0.23 to (1 study) Moderate1
contact with 0.78)
mental health
service) - end of
treatment
Global state 519 per 337 per 1000 RR 0.65 172 ⊕⊕⊕⊝
(relapse, full or 1000 (239 to 482) (0.46 to (2 studies) Moderate1
partial) - end of 0.93)
treatment
Global state 318 per 210 per 1000 RR 0.66 181 ⊕⊕⊝⊝
(remission, full or 1000 (102 to 442) (0.32 to (2 studies) Low1,2
partial) - end of 1.39)
treatment
Global state - N/A Mean global state (functioning/ disability N/A 467 ⊕⊝⊝⊝
functioning / [GAF], end of treatment) in the intervention (2 studies) Very
disability (GAF) - groups was 0.32 standard deviations lower low1,2,3
end of treatment (0.51 to 0.14 lower)
Global state - N/A Mean global state (functioning/ disability N/A 301 ⊕⊕⊕⊝
functioning / [GAF], >12 months’ follow-up) in the (1 study) Moderate1
disability (GAF) - intervention groups was 0.07 standard
>12 months’ deviations lower (0.29 lower to 0.16 higher)
follow-up
Total symptoms N/A Mean total symptoms (panss), end of treatment N/A 99 ⊕⊕⊝⊝
(PANSS) - end of in the intervention groups was 0.52 standard (1 study) Low1,3
treatment deviations lower (0.92 to 0.11 lower)
Positive N/A Mean positive symptoms (PANSS or SAPS, end N/A 468 ⊕⊕⊝⊝
symptoms of treatment) in the intervention groups was (2 studies) Low1,3
(PANSS or 0.21 standard deviations lower (0.39 to 0.03
SAPS) - end of lower)
treatment
Negative N/A Mean negative symptoms (PANSS or SANS, N/A 468 ⊕⊕⊝⊝
symptoms end of treatment) in the intervention groups (2 studies) Low1,3
(PANSS or was 0.39 standard deviations lower (0.57 to 0.2
SANS) - end of lower)
treatment
Positive N/A Mean positive symptoms (PANSS, >12 months’ N/A 301 ⊕⊕⊕⊝
symptoms follow-up) in the intervention groups was 0.06 (1 study) Moderate1
(PANSS) - >12 standard deviations higher (0.16 lower to 0.29
months’ follow-up higher)
Negative N/A Mean negative symptoms (PANSS, >12 months’ N/A 301 ⊕⊕⊕⊝
symptoms follow-up) in the intervention groups was 0.07 (1 study) Moderate1
(PANSS) - >12 standard deviations lower (0.29 lower to 0.16
months’ follow-up higher)
Employment and 347 per 250 per 1000 RR 0.72 436 ⊕⊕⊕⊝
education - end of 1000 (187 to 337) (0.54 to (1 study) Moderate1
treatment 0.97)
Employment and 544 per 577 per 1000 RR 1.06 547 ⊕⊕⊕⊝
education - >12 1000 (501 to 669) (0.92 to (1 study) Moderate1
months’ follow-up 1.23)
Note. CI = confidence interval; RR = risk ratio; GAF = Global Assessment of Functioning; PANSS = Positive and
Negative Syndrome Scale; SANS = Scale for the Assessment of Negative Symptoms; SAPS = Scale for the Assessment
of Positive Symptoms.

2 Evidence of serious heterogeneity of study effect size. 3 Suspicion of publication bias.
Moderate quality evidence from up to three trials (N = 733) showed that EIS was
more effective than standard care in reducing hospitalisation, number of admissions,
number of bed days, and contact with services at the end of the intervention. Two
trials with 467 participants presented very low quality evidence showing a
significant positive effect of EIS on functioning at the end of the intervention.
Moderate to low quality evidence from up to two trials (N = 181) showed that EIS
significantly reduce relapse and have a beneficial effect on psychosis symptoms
(total, positive and negative) at the end of the intervention. There was, however, no
effect on remission (k = 2; N = 181)
One trial (N = 436) presented moderate quality evidence that those receiving EIS
were significantly more likely to be in work or employment at the end of the
intervention.
However, at follow-up exceeding 12 months, there was no evidence of any positive

effects on either critical or non-critical outcomes. No data were available for carer
satisfaction or DUP.
Clinical evidence summary

Overall, the evidence suggests that EIS is effective across all service, clinical and
social outcomes at post-treatment. However, there is no evidence that these positive
effects are maintained at follow-up 12 months after leaving EIS.
Health economics evidence

The systematic literature search identified six economic studies that assessed EIS for
individuals with psychosis and schizophrenia (Cocchi et al., 2011; Hastrup et al.,
2013; McCrone et al., 2010; McCrone et al., 2009d; Mihalopoulos et al., 2009; Serretti
et al., 2009). Both studies by McCrone and colleagues were undertaken in the UK
(McCrone et al., 2010; McCrone et al., 2009d), two studies were undertaken in Italy
(Cocchi et al., 2011; Serretti et al., 2009), one in Denmark (Hastrup et al., 2013) and
one in Australia (Mihalopoulos et al., 2009). Details on the methods used for the
systematic search of the economic literature are described in Chapter 3. References to
included studies and evidence tables for all economic studies included in the
guideline systematic literature review are presented in Appendix 19. Completed
methodology checklists of the studies are provided in Appendix 18. Economic
evidence profiles of studies considered during guideline development (that is,
studies that fully or partly met the applicability and quality criteria) are presented in
Appendix 17, accompanying the respective GRADE clinical evidence profiles.
McCrone and colleagues (2010) evaluated the cost effectiveness of EIS compared
with standard care, defined as care by CMHTs, for 144 service users with psychosis.
This was an economic evaluation undertaken alongside an RCT (CRAIG2004B)

conducted in the UK. The time horizon of the analysis was 18 months and the
perspective of public sector payer was adopted. The study estimated NHS costs
(primary, secondary and community care) and criminal justice costs incurred by
arrests, court appearances and probation. The authors stratified costs, which enabled
them to estimate costs from the NHS and PSS perspectives too. The resource use
estimates were based on the RCT, hospital administrative system records, prison
service annual reports and accounts, and other published sources. The unit costs
were obtained from national sources. The measure of outcome for the economic
analysis was improvement in Manchester Short Assessment of Quality of Life
(MANSA) scores and vocational recovery. Vocational recovery was defined as a
return to or taking up full-time independent employment or full-time education. EIS
resulted in greater improvement in MANSA scores (p = 0.025) and also in a greater
proportion of service users achieving vocational recovery, although the latter
outcome was not statistically significant. The mean cost per person over 18 months
was £11,685 for EIS and £14,062 for standard care in 2003/04 prices, and excluding
criminal justice sector costs the mean cost per person over 18 months was £11,682 for
EIS and £14,034 for standard care. In both cases the cost difference was not
statistically significant possibly because of the low number of participants in the
study. Also, it was found at willingness to pay of £0 for someone making a
vocational recovery the probability of EIS being cost effective is 0.76, and at
willingness to pay of £0 for a unit difference in MANSA scores the probability of EIS
being cost effective is 0.92. Results suggest that EIS provides better outcome at no
extra cost, and thus is a cost-effective intervention for people with psychosis in the
UK. The analysis was judged by the GDG to be directly applicable to this guideline
review and the NICE reference case. The estimate of relative treatment effect was
obtained from a single small RCT and some of the resource use estimates were
derived from local sources, which may limit the generalisability of the findings.
Also, the time frame of the analysis was under 2 years, which may not be sufficiently
long enough to reflect all important differences in costs and clinical outcomes.
Moreover, QALYs were not used, however in this case it was not a problem since the
intervention was found to be dominant. Overall, given the limited availability of
data this was a well-conducted study and was judged by the GDG to have only
minor methodological limitations.
Another study by McCrone and colleagues (2009d) was a model-based cost analysis
that compared EIS with standard care in service users with first episode psychosis.
The authors stated that they were performing a cost-minimisation analysis, however
this assumption was solely based on the authors’ views that intervening early was
unlikely to result in poorer health. Consequently, this was treated as a cost-analysis
in the guideline systematic review. Standard care was defined as any specialised
mental health provision that did not offer any intervention specifically intended to
treat first episode psychosis. The analysis considered costs from the NHS and PSS
perspectives and included costs associated with inpatient, outpatient and
community care. Costs were reported for years one and three. It was found that EIS
resulted in cost savings of £4,972 and £14,248 in years one and three, respectively (in
2006/07 prices). Overall the analysis was judged by the GDG to be directly

applicable to this guideline review and the NICE reference case. Probabilities of
admissions, readmissions and transitioning along care pathways were derived from
a single RCT, local audit data, routine data collected by the Department of Health
and expert judgement; costs for the model were largely obtained from a single RCT,
PSSRU and authors’ assumptions; the definition of standard care was based on
authors’ assumptions and practice described in a single RCT. Nevertheless, the
authors conducted a range of deterministic sensitivity analyses that indicated that
when varying the model’s assumptions EIS costs never exceed the costs of standard
care. Also, probabilistic sensitivity analysis indicated that there is a far greater
likelihood of cost savings associated with EIS and the results were fairly robust. The
analysis was judged by the GDG to have only minor methodological limitations.
Two further studies (Cocchi et al., 2011; Serretti et al., 2009) conducted in Italy
reported similar findings. Cocchi and colleagues (2011) evaluated the cost
effectiveness of EIS compared with standard care (defined as any specialised mental
health provision not offering interventions specifically aimed at treating the first
episode psychosis). The analysis was based on two small cohort studies each with (n
= 23) service users with schizophrenia and related disorders. The analysis was
performed from the Italian NHS perspective and the primary outcome measure was
improvement on the Health of the Nation Outcome Scale (HoNOS). Over the 5 years
EIS resulted in cost savings and greater improvement on the HoNOS scale.
However, the type of treatment did not produce a significant effect on HoNOS
scores at the 5-year follow-up. The study was judged by the GDG to be partially
applicable to this guideline review and the NICE reference case. The findings are
based on a very small sample; and also cohort studies are prone to errors and bias.
Moreover, the unit costs of resource use were obtained from previous publications
and other local sources. Consequently, this analysis was judged by the GDG to have
potentially serious methodological limitations. Similarly, a model-based cost
analysis from the perspective of the Italian NHS by Seretti and colleagues (2009)
compared EIS with standard care in service users with schizophrenia. Standard care
was defined as care provided by community mental health centres. It was concluded
that in year one EIS was a cost-saving strategy. The analysis was judged by the GDG
to be only partially applicable to this guideline review and the NICE reference case.
In the analysis the efficacy data were based on various published sources. The
resource utilisation associated with the standard care was derived from a
retrospective prevalence-based multi-centre study and the resource utilisation
associated with the intervention was based on various published sources and
authors’ assumptions. Moreover the source of unit costs was unclear. For these
reasons the analysis was judged by the GDG to have potentially serious
A recent cost-effectiveness analysis by Hastrup and colleagues (2013) based on a

large RCT (PETERSEN2005) (n = 547) compared EIS with care provided by
community mental health centres in service users with schizophrenia spectrum
disorders from the public sector payer perspective. The mean total costs over 5 years
were lower in the intervention group and the mean GAF score was higher, although

the differences were not statistically significant. Moreover, the probability EIS is cost
effective at willingness to pay of €0 for an extra point increase on the GAF scale was
estimated to be 0.953 and at willingness to pay of €2,000 it was 0.97. The study was
judged by the GDG to be partially applicable to this guideline review and the NICE
reference case. In the analysis, the estimate of relative treatment effect was derived
from a single RCT based in Denmark; the estimates of the resource use were derived
from the same RCT and national registers; the unit cost estimates were from national
and local sources. The study may have limited generalisability to the NHS, but
overall the analysis was well conducted and was judged by the GDG to have only
Similarly in Australia, Mihalopoulos and colleagues (2009) compared EIS with

standard care in service users with schizophrenia, bipolar disorder, depression with
psychotic features, delusional disorder and psychosis. Standard care was defined as
local inpatient and community-based care and the analysis was based on a small
cohort study with historical controls (n = 65). According to the analysis, EIS resulted
in significant annual cost savings from the public mental health service sector
perspective and there were significantly greater improvements on the Brief
Psychiatric Rating Scale (BPRS) during the long-term follow-up of up to 7.2 years. As
a result EIS was identified as a dominant strategy. This study was judged by the
GDG to be partially applicable to this guideline review and the NICE reference case.
The findings are based on a small cohort study with historical controls. Also, the
resource use estimates were derived from a variety of sources including clinical
records, cohort study and other various nationwide sources and as a result findings
may have limited generalisability to the NHS. For these reasons the analysis was
judged by the GDG to have potentially serious methodological limitations.
12.3.3 Early detection programmes to reduce the duration of untreated

psychosis
Introduction
Long DUP is associated with poor clinical outcomes for people with first episode
psychosis (Marshall et al., 2005; Perkins et al., 2005) and poorer quality of life at first
contact with services (Marshall et al., 2005). DUP of months or even years is common
(Marshall et al., 2005; Norman et al., 2006); delays initiating help-seeking and slow
health service response contribute to treatment delay (Malla et al., 2006). In UK
government guidance (Care Services Improvement Partnership, 2005; Department of
Health, 2001), and internationally (Bertolote & McGorry, 2005), professionals within
EIS have been directed to ensure prompt access to treatment for people with first
episode psychosis. Effective means to achieve this, however, are unclear.
Definition and aim of intervention/service system

This review assesses the evidence for the effectiveness of early detection
programmes, that is, any programme designed to reduce DUP and facilitate prompt
access to treatment for people with first episode psychosis.

Clinical review protocol (early detection programmes)
guideline, can be found in Table 135 (the full review protocols and a complete list of
Table 135: Clinical review protocol summary for the review of early detection
programmes to reduce DUP
Review question(s) Are early detection programmes effective in reducing duration of untreated
psychosis and improving pathways to care for people with first episode
psychosis?
Population People with first episode psychosis
Intervention(s) Included
Early detection programmes designed to facilitate access to treatment for first
episode psychosis (involving service reconfiguration and/or public education
campaigns targeting health professionals, other community professionals,
potential service users, or the public).
Excluded
This review was limited to early detection programmes designed to facilitate
access to services and reduce duration of untreated psychosis for people with
first episode psychosis. Psychosis prevention services for people with
prodromal symptoms or at ultra-high risk of psychosis were excluded
Comparison Treatment as usual without early detection programme
Critical outcomes • DUP
• Number of people with first episode psychosis accepted to services
• Health status, experience of care, or referral pathways of people with
first episode psychosis at admission to services.
• Referral behaviours of groups targeted in early detection programmes
databases Process
Topic specific: CINAHL, PsycINFO, IBSS
Date searched 2009 to June 2013 (update search)
Study design Included studies
Any study providing quantitative comparison of an early detection programme
and treatment as usual (in EIS or other mental health services) – that is, cluster
randomised trials, two-group non-randomised comparison studies, pre-post
comparison studies.
Review strategy
Narrative synthesis of the included studies.
Studies considered
The GDG selected an existing systematic review (Lloyd-Evans et al., 2011) as the
basis for this section of the guideline, with a new search conducted to update the

existing review. The review by Lloyd-Evans and colleagues included 11 studies
evaluating eight early detection programmes: LEOCAT 42 (Power et al., 2007),
REDIRECT 43 (Lester et al., 2009b), DETECT 44 (Renwick et al., 2008), EPPIC1 45
(McGorry et al., 1996; Yung et al., 2003), TIPS 46 (Joa et al., 2008; Johannessen et al.,
2001; Melle et al., 2004), EPPIC2 47 (Krstev et al., 2004), EPIP 48 (Chong et al., 2005),
PEPP 49 (Malla et al., 2005).
Two studies of two additional initiatives were identified by the guideline search:
EASY 50 (Chen et al., 2011) and an untitled public education campaign (Yoshii et al.,
2011).
In total, 13 studies of 10 early detection programmes met the eligibility criteria for
this review. All were published in peer-reviewed journals between 1996 and 2012.
Further information about both included and excluded studies can be found in
Lloyd-Evans et al. (2011).
Of the 10 early detection programmes, five evaluated multi-focus public awareness

campaigns (TIPS, EPPIC2, EPIP, PEPP, EASY), three evaluated GP education
programmes (LEOCAT, REDIRECT, DETECT), one evaluated a specialist EIS
(EPPIC1) and one evaluated an online education campaign for parents of high school
students (Untitled; Yoshii et al., 2011). For a full description of the characteristics of
the included and excluded studies, see Lloyd-Evans et al. (2011).
The studies included in this review employed varied study designs. Therefore, a
meta-analysis of the included studies was not conducted and a narrative summary
of the findings is provided below.
Clinical evidence for the review of early detection programmes verses any
control
Significant reductions in mean or median DUP were reported for two out of five
multi-focus public awareness campaigns. The Norwegian TIPS programme reported
a reduction in median DUP from 16 to 5 weeks. The Singapore EPIP programme
reported reductions in mean DUP from 32 to 13 months and in median DUP from 12
to 4 months. Three multi-focus campaigns made no significant difference to DUP.
Two GP education campaigns and one introduction of an EIS led to no significant
reduction in DUP.
42 Lambeth Early Onset Crisis Assessment Team.

43BiRmingham Early Detection In untREated psyChosis Trial.
44Dublin East Treatment and Early Care Team.
45Early Psychosis Prevention and Intervention Centre (1).
46Treatment and Intervention in Psychosis.
47Early Psychosis Prevention and Intervention Centre (2).
48 Early Psychosis Intervention Program.
49 Prevention and Early Intervention in Psychosis Program.
50Early Assessment Service for Young People with Psychosis program.

No clear effect was observed in the number of people with first episode psychosis
referred to services following an early detection programme. Studies of multi-focus
public awareness programmes and a GP education programme reported no
significant change in number of new referrals accepted.
Four studies evaluated pathways to care. For one GP education programme, and one
multi-focus public awareness programme, no significant difference with comparison
groups was found in the source of the referral. However, one UK GP education
programme found that patients from GP practices receiving the intervention were
less likely to have contact with A&E departments in their pathway to mental health
services. One multi-focus public awareness programme reported that during the
campaign, people were significantly more likely to self-refer and less likely to be
referred via the police than in the historical comparison period.
People from areas exposed to a multi-focus public awareness programme were

found to have significantly less severe symptoms at first contact with services than
those from comparison groups in the Norwegian TIPS Project and the Australian
EPPIC programme. No significant difference in service users’ symptom severity was
found between intervention and comparison areas in the Canadian multi-focus
public awareness programme. The REDIRECT study found no significant difference
in symptom severity or premorbid adjustment between people admitted from areas
included in a GP education campaign and comparison areas.
All three studies of GP education initiatives included in this review found some
evidence of impact of the initiative on GPs’ referral behaviour. DETECT and
LEOCAT reported that GPs receiving education were more likely to refer people
with first episode psychosis to mental health services than GPs in a comparison
group. REDIRECT found that the time from service users’ first contact with GPs to
referral to EIS was significantly shorter in duration for people from GP surgeries in
the intervention arm of the study. One study reported a significant increase in help-
seeking behaviour in parents of junior and high school students following a web-
based educational programme. No change in DUP or number of referrals resulting
from changes in referrers’ behaviour was demonstrated in any of these studies.

GP education programmes and setting up specialist EIS by themselves had no
impact on DUP. Overall, there is no compelling evidence that any types of early
detection programme are effective in reducing DUP or increasing numbers of people
with first episode psychosis presenting to services.
12.3.4 Community mental health teams

Introduction
One of the earliest service developments in community-based care was that of the
community mental health team (CMHT) (Merson et al., 1992). CMHTs are
multidisciplinary teams, comprising all the main professions involved in mental

health, including psychiatry, psychology, nursing, occupational therapy and social
work. Having developed in a relatively pragmatic way, CMHTs became the
mainstay of community-based mental health work in most developed countries
(Bennett & Freeman, 1991; Bouras et al., 1986), as well as in many others (Isaac, 1996;
Pierides, 1994; Slade et al., 1995). Nevertheless, concerns about CMHTs have been
raised, particularly regarding the incidence of violence (Coid, 1994), the quality of
day-to-day life for people with serious mental illness and their carers, and the impact
on society (Dowell & Ciarlo, 1983). In addition, CMHTs have changed very
considerably over time in terms of how they are configured, what they provide, their
role and their integration within the wider systems of mental health and social care.
Definition and aim of intervention/service system
The GDG judged that the definitions used for the first (2002) guideline for CMHTs
and the comparator standard care or usual care were still applicable:
• **2002**CMHT care was management of care from a multidisciplinary,

community-based team (that is, more than a single person designated to work
within a team)
• standard care or usual care must be stated to be the normal care in the area
concerned, non-team community care, outpatient care, admission to hospital
(where acutely ill people were diverted from admission and allocated to
CMHT or inpatient care) or day hospital care.**2002**
The review specifically focused upon CMHT management, and therefore excluded
studies that involved any additional method of management in the CMHT.
Clinical review protocol (community mental health teams)

guideline can be found in Table 136 (the full review protocols and a complete list of

Table 136: Clinical review protocol summary for the review of community mental
health teams
potential harms of community mental health teams compared with treatment
as usual or another intervention?
Objectives To evaluate the clinical effectiveness of community mental health teams in the
Intervention(s) Community mental health teams
Critical outcomes • Service use
o Hospitalisation: mean number of days per month in hospital
o Not remaining in contact with psychiatric services
o Use of services outside of mental health provision (that is,
emergency services)
• Employment status
• Accommodation status
• Quality of life
• Mental state
o General symptoms
o Total symptoms
o Positive symptoms
o Negative symptoms
• Satisfaction
o Participant satisfaction
o Carer satisfaction
Process
Date searched SR/RCT: 2002 to June 2013
Study design RCT
point reported within the time-point groupings
Sub-analysis
schizophrenia/schizoaffective disorder or psychosis.

studies.

Three RCTs (N = 344) met the eligibility criteria for this review: GATER1997 (Gater
et al., 1997), MERSON1992 (Merson et al., 1992), TYRER1998 (Tyrer et al., 1998). The
included trials were published between 1992 and 1998. All were conducted in the
UK. Further information about both included and excluded studies can be found in
Appendix 15a.
Of the included trials, two involved a comparison of a CMHT with standard hospital
treatment and one compared CMHTs with traditional psychiatric services. The
proportion of individuals with psychosis and schizophrenia ranged from 38 to 100%.
The length of follow-up ranged from 12 to 104 weeks. Table 137 provides an
overview of the included trials.
This review did not combine data from the three included trials in statistical
analysis. MERSON1992 and TYRER1998 could not be combined in meta-analysis
because in TYRER1998 the service was seeing discharged psychiatric patients who,
presumably, were more likely to be readmitted to hospital and be more severely ill
than those in the other two trials. This would appear to be confirmed by the very
high admission rates in TYRER1998. Further, GATER1997 could not be included in
meta-analysis due to the possibility of unit of analysis error as the study used a
cluster randomisation design and there is no indication of accounting for inter-class-
correlation. Further information about the cluster design was requested from the
authors. The findings from all three included trials are thus described narratively.
Table 137: Study information table for trials of community mental health teams
versus standard care
Community mental health teams versus standard care

Study ID(s) GATER1997
MERSON1992
TYRER1998
Country UK (k = 3)
Mean age of participants (range) 38.07 years (32 to 44.13 years)1
Mean percentage of participants with primary 64.49% (38% to 100%)
Mean gender % women (range) 50.79% (41.57 to 60%)1
Intervention type Community focused multidisciplinary team (EIS) (k = 1)
Community team (k = 2)
Comparisons Standard hospital treatment (k = 2)
Traditional psychiatric service (k = 1)
Note.1 TYRER1998 did not report data.

Clinical evidence for community mental health teams
Two trials (MERSON1992, TYRER1998) reported that CMHTs did not have a
significant benefit over standard care on the number of participants admitted to
hospital, use of A&E services, contact with primary care or contact with social care at
both short- and medium-term follow-up. Additionally, one study (GATER1997) did
not find any difference between CMHTs and standard care in the number of
participants in contact with mental health services at medium-term follow-up. There
was no significant difference between groups in psychological health and social
functioning (MERSON1992). No study reported data for quality of life, mental state
or satisfaction.

Despite the fact that CMHTs became the mainstay of community mental healthcare,
there is surprisingly little evidence to show that they are an effective way of
organising services. Moreover, the trials of CMHTs included here are very unlikely
to reflect the enormous diversity of community mental healthcare today, most of
which has absorbed the practices used by more recently developed services such as
ACT, outreach services, ICM and even early intervention. As such, evidence
presented here for or against the effectiveness of CMHTs in the management of
psychosis and schizophrenia is insufficient to make any evidence-based
recommendations.

The systematic search of the economic literature undertaken for the 2014 guideline,
identified only one eligible study on CMHTs for individuals with psychosis and
schizophrenia (McCrone et al., 2010). Details on the methods used for the systematic
search of the economic literature are described in Chapter 3. References to included
studies and evidence tables for all economic studies included in the guideline
systematic literature review are presented in Appendix 19. Completed methodology
checklists of the studies are provided in Appendix 18. Economic evidence profiles of
studies considered during guideline development (that is, studies that fully or partly
met the applicability and quality criteria) are presented in Appendix 17,
accompanying the respective GRADE clinical evidence profiles.
McCrone and colleagues (2010) evaluated the cost effectiveness of CMHTs compared
with EIS for 144 service users with psychosis. This was an economic evaluation
based on an RCT (CRAIG2004B) conducted in the UK. The time horizon of the
analysis was 18 months and the public sector payer perspective was adopted,
although the authors reported stratified costs and this allowed estimation of costs
from the NHS and PSS perspective. CMHTs resulted in lower quality of life scores
on the MANSA scale (p = 0.025) and fewer service users achieving vocational
recovery (p = ns) compared with EIS. The mean cost per person over 18 months was
£14,062 for CMHTs and £11,685 for EIS in 2003/04 prices, and excluding criminal
justice sector costs the mean cost per person over 18 months was £14,034 for CMHTs
and £11,682 for EIS. In both cases the cost difference was not statistically significant
possibly because of the low number of participants in the study. Results suggest that
CMHTs lead to worse health outcomes and potentially higher healthcare costs.
Consequently, EIS is a preferred treatment strategy compared with CMHTs. For
more details and discussion of the findings see Section 12.3.2.
12.3.5 Intensive case management

Introduction
In existence for at least 40 years, assertive community treatment (ACT) and intensive
case management (ICM) are approaches to caring for people with severe mental
illness (typically schizophrenia or bipolar disorder) who require intensive
community support and have frequent admissions. Although in the 2002 and 2009
guidelines, these interventions were treated as discrete approaches, for the purposes
of the 2014 guideline they are considered together as they are similar: both use an
assertive outreach model of care (that is, persisting with service users who are not
engaging) and both specify that practitioners should carry limited caseloads. The
main difference is that ACT requires team members to share responsibility for the
teams’ clients, whereas ICM puts greater emphasis on the primacy of the individual
case manager. A further difference is that ACT has been more precisely defined than
ICM, for example, in terms of requirements for daily team meetings and for certain
professionals to be included in the team, but has also become less distinct from it, as
case managers have increasingly adopted a team-based approach and other elements
of the ACT model (Marshall, 2008).
Early Cochrane reviews in this area attempted to draw a categorical distinction

between trials of ACT and ICM on the basis of the label that the trialists had given to
the intervention (Marshall et al., 2000; Marshall & Lockwood, 2000). ACT and ICM
trials were then analysed in separate meta-analyses. However, it became
increasingly obvious that such labels bore little relationship to actual practice in the
trial. Later reviews, including Cochrane reviews (Burns et al., 2007b; Dieterich et al.,
2010), therefore abandoned this blunt categorical approach and instead obtained
ratings of fidelity to the ACT model for ACT and ICM interventions, based on data
obtained directly from trialists. Trials of ACT and ICM were then combined in the
same meta-analysis and fidelity to the ACT model used as an explanatory covariate
whenever outcomes showed significant heterogeneity. The GDG accepts this
approach, which has a sounder empirical basis than earlier reviews and takes
account of the complexity of the changes in community care over time and across
countries.
The question of control conditions is also problematic because standard care has
been evolving from a clinic-based approach to a team-based community model,
incorporating strong elements of case management (such as the UK care programme
approach [CPA]). In accordance with the most recent Cochrane review (Dieterich et
al., 2010), the GDG has distinguished two types of control: standard care, which
refers to a clinic-based approach to follow-up; and non-intensive case management,
which refers to a case management approach to follow-up, where the caseload size is
large.

The definitions used in this review for ICM, non-intensive case management (non-
ICM) and standard care used in the Cochrane review (Dieterich et al., 2010) and
adopted for this guideline, are as follows:
ICM:
Where the majority of people received a package of care shaped either on:
• the ACT model, being based on the Training in Community Living project
and the Program of Assertive Community Treatment (PACT) (Stein & Test,
1980), or
• the assertive outreach model (Witheridge, 1991; Witheridge et al., 1982), that
is, a multidisciplinary team-based approach, practicing ’assertive outreach’
and providing 24 hours’ emergency cover (McGrew & Bond, 1995), or
• the case management model (Intagliata, 1982) however it was described in the
trial
• report with a caseload up to and including 20 people.
Non-ICM: Where the majority of people received the same package of care as
described for ICM (above) but with a caseload of over 20 people.
Standard care: Where the majority of people received a community or outpatient

model of care not specifically shaped on either the model of ACT and case
management, and not working within a specific designated named package or
approach to care.
Clinical review protocol (intensive case management)


Table 138: Clinical review protocol summary for the review of intensive case
management
potential harms of intensive case management compared with non-intensive
case management or standard treatment?
Objectives To evaluate the clinical effectiveness of intensive case management in the
treatment of psychosis and schizophrenia
Intervention(s) Intensive case management
Comparison i) Non-intensive case management
ii) Standard care
emergency services)
• Quality of life
• Satisfaction
databases Process
Topic specific: CINAHL, PsycINFO,
Study design RCTs
Sub-analysis
Where data were available, sub-analyses were conducted of studies with >75%
of the sample described as having a primary diagnosis of schizophrenia/
Where data were available, sub-analyses were conducted for UK only studies.

The GDG selected an existing Cochrane review (Dieterich et al., 2010) as the basis for
this section of the guideline, with a new search conducted to update it. The Cochrane
review included 38 RCTs (N = 7,328) that met eligibility criteria for this guideline:
Aberg-Wistedt-Sweden (Aberg-Wistedt et al., 1995), Audini-UK (Audini et al., 1994),
Bjorkman-Sweden (Bjorkman et al., 2002), Bond-Chicago1 (Bond et al., 1990), Bond-
Indiana1 (Bond et al., 1988), Bush-Georgia (Bush et al., 1990), Chandler-California1
(Chandler et al., 1996), Curtis-New York (Curtis et al., 1992), Drake-NHamp (Drake
& McHugo, 1998), Essock-Connecticut1 (Essock & Kontos, 1995), Essock-
Connecticut2 (Essock et al., 2006), Ford-UK (Ford et al., 1995), Hampton-Illinois
(Hampton et al., 1992), Harrison-Read-UK (Harrison-Read et al., 2002), Herinckx-
Oregon (Herinckx et al., 1997), Holloway-UK (Holloway & Carson, 1998), Jerrell-
SCarolina1 (Jerrell, 1995), Johnston-Australia (Johnston et al., 1998), Lehman-
Maryland1 (Lehman et al., 1997), Macias-Utah (Macias et al., 1994), Marshall-UK
(Marshall et al., 1995), McDonel-Indiana (McDonel et al., 1997), Morse-Missouri1
(Morse et al., 1992), Morse-Missouri3 (Morse et al., 2006), Muijen-UK2 (McCrone et
al., 1994), Muller-Clemm-Canada (Muller-Clemm, 1996), Okpaku-Tennessee
(Okpaku & Anderson, 1997), OPUS-Denmark (Jørgensen et al., 2000), Pique-
California (Pique, 1999), Quinlivan-California (Quinlivan et al., 1995), REACT-UK
(Killaspy et al., 2006), Rosenheck-USA (Rosenheck et al., 1993), Salkever- SCarolina
(Salkever et al., 1999), Shern-USA1 (Shern et al., 2000), Solomon-Pennsylvania
(Solomon et al., 1994), Sytema-Netherlands (Sytema et al., 2007), Test-Wisconsin
(Test et al., 1991), UK-700-UK (Burns et al., 1999). No additional RCTs were
identified by the guideline search. All 38 studies were published in peer-reviewed
journals between 1988 and 2007. Further information about included studies can be
found in Appendix 15a. Further information about excluded studies can be found in
Dieterich et al. (2010).
All included trials included sufficient data to be included in the meta-analysis. Of

the 38 included trials, 26 trials compared ICM with standard care, 11 trials compared
ICM with non-ICM and one study evaluated both comparisons. Table 139 provides
an overview of the trials included in each comparison.
Two sub-analyses were conducted. The first used 13 trials with a large proportion
(≥75%) of participants with a primary diagnosis of psychosis or schizophrenia. The
second analyses included UK only based trials (k = 8).

Table 139: Study information table for trials comparing ICM with standard care and ICM with non-ICM
ICM versus standard care ICM versus non-ICM

Total no. of trials (k); participants k = 27; N = 4865 k = 12; N = 2560
(N)
Study ID(s) Aberg-Wistedt-Sweden Bush-Georgia
Audini-UK Drake-NHamp
Bjorkman- Sweden Essock-Connecticut1
Bond-Chicago1 Essock-Connecticut2
Bond-Indiana1 Harrison-Read-UK
Chandler-California1 Johnston-Australia
Curtis-New York McDonel-Indiana
Ford-UK Okpaku-Tennessee
Hampton-Illinois Quinlivan-California
Herinckx-Oregon REACT-UK
Holloway-UK Salkever- SCarolina
Jerrell-SCarolina1 UK-700-UK
Lehman-Maryland1
Macias-Utah
Marshall-UK
Morse-Missouri1
Morse-Missouri3
Muijen-UK2
Muller-Clemm-Canada
OPUS-Denmark
Pique-California
Quinlivan-California
Rosenheck-USA
Shern-USA1
Solomon-Pennsylvania
Sytema-Netherlands
Test-Wisconsin
Country Canada (k = 1) Australia (k = 1)
Denmark (k = 1) UK (k = 3)
Netherlands (k = 1) USA (k = 8)

Sweden (k = 2)
UK (k = 5)
USA (k = 17)
Mean age of participants (range) 37.14 years (23 to 48 years)1 37.81 years (34 to 41.54 years)4
Mean percentage of participants with 67.36% (30 to 100%)2 69.67% (23 to 88.89%)
Mean gender % women (range) 37.34% (0 to 59%)3 42.24% (25.6 to 57%)
Length of follow-up (range) 26 to 156 weeks 17 to 156 weeks
Intervention type ACT according to the Stein & Test model (k = 15) Employment oriented case management (k = 1)
ACT according to the Stein & Test model staffed by consumers ACT according to the Stein & Test model (k = 3)
(k = 1) Clinical case management according to the Stein & Test
Case management approach provided by a community model (Training in Community Living programme) (k
support team (k = 1) = 2)
Case management based on the Strength Model (k = 2) Generalist model of ACT (k = 1)
Case management from team of social service case managers (k Enhanced community management based on ACT
= 1) principles (Stein model) (k = 1)
Choices programme (k = 1) ACT teams with special training in substance misuse
Clinical case management based on ACT principles (Training treatment (k = 1)
in Community Living programme) (k = 2) ACT (McGrew & Bond, 1995) (k = 1)
ICM according to the ‘Clinical Case Management Model’ PACT (k = 1)
developed by Kanter (k = 1) ICM (k = 1)
ICM (not following any specific model of case management) (k
= 1)
ICM provided by an individual forensic case manager (k = 1)
Intensive broker case management Model (k = 1)
Intensive outreach case management (k = 1)
Modified ACT (k = 1)
Programme assertive community treatment (PACT) adaptation
(k = 1)
Comparisons Psychosocial rehabilitation programme (k = 1) Standard case management from a community mental
Routine care from psychiatric services (k = 6) health centre (k = 2)
Routine outpatient care (k = 2) Non-ICM provided by the mental health services (k =
Services as usual (k = 6) 1)

Services offered by the public mental health system (k=1) Generalist model, but providing case managers mobile
Standard care provided by CMHTs (k = 6) (k = 1)
Standard care provided by community psychiatric nursing Standard care providing case management at a lower
service (k = 2) level of intensity and rehabilitation services (k = 1)
Standard care provided from a variety of agencies (k = 1) Traditional case management programme (k = 1)
Standard care provided from a drop-in centre (k = 2) Clinical Case Management (k = 2)
Locality-based community psychiatric services (k = 1)
Non-ICM, incorporating most of the ACT principles,
but providing less individual service for substance
misuse (k = 1)
Services offered by CMHT (according to the CPA) (k =
1)
Case management (k = 1)
Note. CMHT = community mental health team; ICM = intensive case management; SC = Standard care; non-ICM = non-intensive case management; CPA = care programme
approach.
1 Chandler-California1, Jerrell-SCarolina1, Macias-Utah, Muller-Clemm-Canada and Pique-California did not report data.
2Pique-California and Shern-USA1 did not report data.
3Pique-California did not report data.
4 Bush-Georgia l did not report data.

Clinical evidence for intensive case management
Intensive case management versus standard care

Low quality evidence from 24 trials (N = 3,595) showed that ICM was more effective
than standard care in reducing the average number of days in hospital per month,
and keeping in contact with psychiatric services at medium- and long-term follow-
up.
Low quality evidence from a single study (N = 125) found a positive effect of ICM on
self-reported quality of life at short-term follow-up. However, this effect was not
found at either medium- or long-term follow-up.
Moderate quality evidence from up to five trials (N = 818) showed that ICM was
more effective than standard care in improving global functioning at both short- and
long-term but not medium-term follow-up.
Very low to high quality evidence from up to two trials (N = 500) showed that
participants receiving ICM were more satisfied with the intervention than those
receiving standard care at all follow-up points.
No studies reported usable data on carer satisfaction.

The sub-analysis of trials with a sample of ≥75% people with psychosis and
schizophrenia upheld the positive effect found in the main analysis of ICM on both
the average number of days in hospital and self-reported quality of life. Consistency
with the main analysis was also found for remaining in contact with psychiatric
services at medium-term follow-up. However, unlike the main analysis, no
significant difference for remaining in contact with psychiatric services was reported
at long-term follow-up. Moreover, no difference between groups was observed for
satisfaction with services at short-term follow-up or for functioning at any follow-up
point. See Appendix 16 for the related forest plots.
Sub-analysis (UK only)

Unlike the main analysis, the UK only sub-analysis found no significant effect of
ICM in reducing the average number of days hospitalised when compared with
standard care (k = 5; N = 369). The sub-analysis findings did not differ from the main
analysis in finding a benefit of ICM on both remaining in contact with psychiatric
services and satisfaction at short-term follow-up, and no effect of ICM on quality of
life. However, unlike the main analysis, participant satisfaction at long-term follow-

was not significantly different between ICM and standard care. No other critical
Table 140: Summary of findings tables for ICM compared with standard care

Intervention: ICM
Comparison: Standard care
Outcomes Illustrative comparative risks* (95% CI) Relative No. of Quality of the
Assumed Corresponding risk effect participants evidence
risk (95% CI) (studies) (GRADE)
Control ICM
Service use N/A Mean service use (average N/A 3,595 ⊕⊕⊝⊝
(average number of days in hospital (24 studies) Low1,2
number of days per month - by about 24
in hospital per months) in the intervention
month) - by groups was 0.86 lower (1.37
about 24 to 0.34 lower)
months
Not remaining Study population RR 0.54 95 ⊕⊝⊝⊝
in contact with 383 per 207 per 1000 (0.28 to 1.05) (1 study) Very low3,4
psychiatric 1000 (107 to 402)
services - short
term follow-up
Not remaining Study population RR 0.51 1,063 ⊕⊕⊕⊝
in contact with 246 per 126 per 1000 (0.36 to 0.71) (3 studies) Moderate1
services -
medium term
follow-up
Not remaining Study population RR 0.27 475 ⊕⊕⊝⊝
in contact with 303 per 82 per 1000 (0.11 to 0.66) (5 studies) Low1,2
services - long
term follow-up
Not remaining Study population RR 0.43 1,633 ⊕⊝⊝⊝
in contact with 270 per 116 per 1000 (0.3 to 0.61) (9 studies) Very low2,5
services - total
Quality of life - N/A Mean quality of life (by short N/A 125 ⊕⊕⊝⊝
by short term term follow-up) in the (1 study) Low4,6
follow-up intervention groups was 0.53
Quality of life - N/A Mean quality of life (by N/A 52 ⊕⊕⊝⊝
by medium medium term follow-up - (1 study) Low4,6
term follow-up LQOLP) in the intervention
(LQoLP) groups was 0.09 lower (0.78
Quality of life - N/A Mean quality of life (by N/A 81 ⊕⊕⊕⊝
by medium medium term follow-up- (1 study) Moderate4
term follow-up MANSA) in the intervention
(MANSA) groups was 0.2 lower (0.69

Quality of life - N/A Mean quality of life (by long N/A 113 ⊕⊕⊝⊝
by long term term follow-up - LQOLP) in (2 studies) Low1,4
follow-up the intervention groups was
(LQoLP) 0.23 higher (0.08 lower to 0.55
higher)
Quality of Life N/A Mean quality of life (by long N/A 132 ⊕⊕⊝⊝
- by long term term follow-up - QOLI) in the (2 studies) Low1,4
(QOLI) 0.09 lower (0.42 lower to 0.24
higher)
Participant N/A Mean participant satisfation N/A 61 ⊕⊝⊝⊝
satisfation - by (by short term follow-up) in (1 study) Very low6,7,8
short term the intervention groups was
follow-up 6.2 lower (9.8 to 2.6 lower)
Participant N/A Mean participant satisfation N/A 500 ⊕⊕⊕⊕
satisfation - by (by medium term follow-up) (2 studies) high
medium term in the intervention groups
follow-up was 1.93 lower (3.01 to 0.86
lower)
Participant N/A Mean participant satisfation N/A 423 ⊕⊕⊕⊝
satisfation - by (by long term follow-up) in (2 studies) Moderate9
long term the intervention groups was
Global N/A Mean global functioning N/A 797 ⊕⊕⊕⊝
functioning (GAF- by short term follow- (4 studies) Moderate1
(GAF) - by up) in the intervention
short term groups was
Global N/A Mean global functioning N/A 722 ⊕⊝⊝⊝
functioning (GAF- by medium term (3 studies) Very low1,2,4
(GAF) - by follow-up) in the intervention
medium term groups was
follow-up 0.09 lower (3.28 lower to 3.11
higher)
Global N/A Mean global functioning N/A 818 ⊕⊕⊕⊝
functioning (GAF- by long term follow- (5 studies) Moderate1
(GAF) - by up) in the intervention
long term groups was
Note. CI = confidence interval; RR = risk ratio; LQoLP = Lancashire Quality of Life Profile; MANSA = Manchester
Short Assessment of Quality of Life; GAF = Global Assessment of Functioning; QOLI = Quality of Life Inventory.
3 Crucial limitation for one or more criteria sufficient to substantially lower confidence in the estimate of effect.
5 Most information is from studies at high risk of bias.
estimate of effect.
7 Concerns regarding applicability - different populations.
9 Concerns regarding size of effect.

Intensive case management versus non-intensive case management
presented in Table 141.The full evidence profiles and associated forest plots can be
Low quality evidence from 12 studies (N = 2,220) showed no difference between

ICM and non-ICM in the average number of days spent in hospital. Further low
quality evidence from a single trial (N = 73) did show a benefit of ICM over non-ICM
in remaining in contact with psychiatric services at medium-term follow-up.
However, this effect was not found at long-term follow-up (k = 3; N = 1,182).
Moreover, there was no difference between ICM and non-ICM in quality of life,
participant satisfaction or global functioning at any follow-up points.
No studies reported usable data on carer satisfaction.

The sub-analysis findings did not differ from the main analysis, reporting no benefit
of ICM over non-ICM for service use outcomes, quality of life, participant
satisfaction or global functioning.
Sub-analysis (UK only)

The sub-analysis findings did not differ from the main analysis reporting no benefit
of ICM over non-ICM for service use outcomes, quality of life, participant
satisfaction or global functioning.
Table 141: Summary of findings tables for ICM compared with non-ICM

Intervention: ICM
Comparison: Non-ICM
Non-ICM ICM (GRADE)
Service use N/A Mean service use (average number of N/A 2,220 ⊕⊕⊝⊝
(average number of days in hospital per month - by about (12 studies) Low1,2
days in hospital per 24 months) in the intervention groups
month) - by about was 0.08 lower (0.37 lower to 0.21
24 months higher)
Not remaining in Study population RR 0.27 73 ⊕⊕⊝⊝
contact with 306 per 82 per 1000 (0.08 to (1 study) Low2,3
psychiatric services 1000 (24 to 266) 0.87)
- medium term
follow-up
Not remaining in Study population RR 0.82 1,182 ⊕⊝⊝⊝
contact with 111 per 91 per 1000 (0.34 to (3 studies) Very
psychiatric services 1000 (38 to 220) 1.98) low1,2,4
- long term

Quality of life - by N/A Mean quality of life (by short term N/A 203 ⊕⊕⊝⊝
short term follow- follow-up) in the intervention groups (1 study) Low2,3
up was 0.02 higher (0.39 lower to 0.43
higher)
Quality of life - by N/A Mean quality of life (by medium term N/A 203 ⊕⊕⊝⊝
medium term follow-up) in the intervention groups (1 study) Low2,3
follow-up was 0.04 higher (0.35 lower to 0.43
higher)
Quality of life N/A Mean quality of life (LQoL - by long N/A 526 ⊕⊕⊕⊝
(LQoLP) - by long term follow-up) in the intervention (1 study) Moderate3
term follow-up groups was
0.03 lower (0.16 lower to 0.1 higher)
Quality of life N/A Mean quality of life (MANSA - by long N/A 166 ⊕⊕⊕⊝
(MANSA) - by term follow-up) in the intervention (1 study) Moderate5
long term follow-up groups was 0.1 lower (0.39 lower to
0.19 higher)
Quality of life N/A Mean quality of life (overall life N/A 203 ⊕⊕⊝⊝
(overall life satisfaction – QOLI - by long term (1 study) Low2,3
satisfaction - follow-up) in the intervention groups
QOLI) - by long was 0.1 lower
term follow-up (0.45 lower to 0.25 higher)
Participant N/A Mean participant satisfaction (patient N/A 585 ⊕⊕⊝⊝
satisfaction (patient need – CAN - by long term follow-up) (1 study) Low2,3
need - CAN)- by in the intervention groups was 0.29
long term follow-up lower
Global functioning N/A Mean global functioning (HONOS - N/A 118 ⊕⊕⊝⊝
(HoNOS) - short short term follow-up) in the (1 study) Low2,3
term follow-up intervention groups was 0.60 higher
(1.8 lower to 3 higher)
Global functioning N/A Mean global functioning (HONOS- N/A 239 ⊕⊕⊝⊝
(HoNOS) - long long term follow-up) in the (1 study) Low2,3
term follow-up intervention groups was 0.40 lower
Note. CI = confidence interval; RR = risk ratio; LQoLP = Lancashire Quality of Life Profile; MANSA = Manchester
Short Assessment of Quality of Life; QOLI = Quality of Life Inventory; CAN = Camberwell Assessment of Need
interview; HoNOS = Health of the Nation Outcomes Scales.
estimate of effect.
5 Optimal information size not met

When compared with standard care worldwide, ICM was found to be effective at
both reducing duration spent in hospital and improving retention in care.
Furthermore, participants consistently reported being more satisfied with the
service. The benefits of ICM on functioning and quality of life are however less
definitive, with inconsistent findings across follow-up points.

Notably, when analysing UK only studies, results did not demonstrate a benefit of
ICM over standard care. The large effect on duration of hospitalisation was no
longer reported and satisfaction data proved inconsistent across time. However, UK
only data do suggest that ICM retains people within the service better than standard
care.
When ICM is compared with a non-ICM intervention, there is inconclusive evidence

about the additional benefits of a more intensive approach to case management.

The economic review identified four eligible studies that met the inclusion criteria
for this guideline. Two studies were conducted in the UK (Harrison-Read et al., 2002;
McCrone et al., 2009c), one study in the US (Slade et al., 2013), one study in Germany
(Karow et al., 2012) and one in Australia (Udechuku et al., 2005). Details on the
Chapter 3. References to included studies and evidence tables for all economic
studies included in the guideline systematic literature review are presented in
Appendix 19. Completed methodology checklists of the studies are provided in
Appendix 18. Economic evidence profiles of studies considered during guideline
development (that is, studies that fully or partly met the applicability and quality
criteria) are presented in Appendix 17, accompanying the respective GRADE clinical
evidence profiles.
The two UK studies were both based on RCTs. Harrison-Read and colleagues (2002)
conducted a cost minimisation analysis comparing ICM, defined as enhanced
community management, versus standard care. Standard care included local
psychiatric services. The authors adopted a cost-minimisation approach since the
effectiveness analysis of trial results found no differences in clinical outcomes. The
study was based on a medium-sized RCT (n = 193) (Harrison-Read-UK) in people
with schizophrenia and related diagnoses. The time horizon of the analysis was 2
years and the NHS and PSS perspective was adopted. The authors considered
inpatient, outpatient and community care costs. In year one ICM resulted in a cost
increase of £441 (p = ns) and in year two in a cost saving of £347 (p = ns) in 1995/96
prices, leading to an overall cost increase of £94 over 2 years. The authors concluded
that ICM did not lead to any important clinical gains or reduced costs of psychiatric
care. Even though the study did not consider QALYs, the authors did not find
differences in clinical outcomes, consequently the study was judged by the GDG to
be directly applicable to this guideline review and the NICE reference case. The
analysis derived some of the unit cost estimates from local sources, which may limit
the generalisability of the findings to the NHS. However, overall this was a well-
conducted analysis with only minor methodological limitations.
McCrone and colleagues (2009c) assessed the cost effectiveness of ICM compared
with standard care. ICM was defined as assertive community management and
standard care as care from CMHTs. The study population comprised service users
with schizophrenia, schizoaffective disorder, bipolar disorder and other psychotic

illnesses. The analysis was based on a relatively large RCT (KILLASPY2006) (n =
251). The time horizon of the analysis was 18 months and the societal perspective
was adopted. However, NHS and PSS costs were reported separately. The analysis
considered: inpatient, outpatient and community care costs; criminal justice costs
incurred by probation, incarceration, lawyers, courts and police; and informal care
costs. The RCT did not find clinical outcomes to be significantly different between
the two groups. However, the authors hypothesised that interventions similar in
effectiveness may differ in terms of process and acceptability. Consequently, the
primary outcome measure of the analysis was satisfaction with services as measured
on Gerber and Prince’s scale. ICM resulted in a cost increase of £3,823 in 2003/04
prices excluding informal care and costs accruing to the criminal justice system.
Including costs from the societal perspective ICM resulted in a cost increase of
£4,031. Cost differences were not statistically significant. Also, it was found that ICM
led to a significantly higher satisfaction score of 79.4 versus 71.7 (p < 0.05) on Gerber
and Prince’s satisfaction scale. As a result, the authors concluded that there was no
difference between the interventions in terms of costs, however ICM resulted in
greater levels of service user satisfaction and engagement, and as such is the
preferred community treatment. However, the cost-effectiveness acceptability curve
showed that for the ICM to be cost effective in 95% of service users, society would
need to be willing to pay £2,500 for one additional unit improvement in the
satisfaction score, which is unlikely to represent ‘good value for money’. Overall the
study was judged by the GDG to be partially applicable to this guideline review and
the NICE reference case. The authors did not attempt to estimate QALYs and the use
of satisfaction scores as an outcome measure made it difficult to interpret the cost
effectiveness results and to compare the findings with other studies. Nevertheless,
this was a well-conducted study and was judged by the GDG to have only minor
A recent cost analysis by Slade and colleagues (Slade et al., 2013) in the US based on
a large observational study (n = 6,030) compared ICM (defined as ACT) with care
without an ACT component. The study population comprised service users with
schizophrenia and bipolar disorder. The analysis was performed from a mental
health service payer perspective and adopted a 1-year time horizon. Mean annual
costs were estimated to be $28,881 versus $27,250 for ICM and standard care groups,
respectively (p = 0.038). The study was judged by the GDG to be only partially
applicable to this guideline review and the NICE reference case. The analysis was
based on a pre- and post-observational study. These studies are prone to bias
because of the inability to control for confounding factors. However, the authors
used an extensive regression approach to control for a range of confounders. Overall
this was a well-conducted cost analysis and was judged by the GDG to have only
A recent cost-utility study by Karow and colleagues (2012) based on a prospective

cohort study (n = 120) in people with schizophrenia spectrum disorders in Germany
compared ICM (defined as ACT) with standard care. Standard care included
inpatient care, care at day clinic and outpatient centre, and care by private

psychiatrists. The public sector payer perspective was adopted and the time horizon
of the analysis was 1 year. The analysis included costs associated with admissions,
outpatient visits, medications and intervention provision. The primary outcome
measure was the QALY. The quality of life was assessed with the EQ-5D descriptive
system and the EQ-5D index scores from the UK were used. ICM resulted in a cost
saving of €2,502 (p = ns) in 2007 prices and an increase in QALYs of 0.1 (p < 0.01) at 1
year’s follow-up. Consequently, ICM was found to be the dominant strategy. Also,
the probability of ICM being cost effective at a willingness to pay of €50,000 per
QALY gained was estimated to be 0.995. The analysis was based on a relatively small
cohort study and was judged by the GDG to be only partially applicable to this
guideline review and the NICE reference case because it was conducted in Germany
and the definition of standard care was very different from the UK. Nevertheless this
was a well-conducted study and was judged by the GDG to have only minor
A cost analysis by Udechuku and colleagues (2005) in Australia based on a pre- and
post-observational study (n = 31) found ICM (defined as ACT) to be cost saving
when compared with care without an ACT component. The study population
comprised service users with schizophrenia, schizoaffective disorder and bipolar
affective disorder. The analysis was performed from the mental health service payer
perspective and adopted a 1-year time horizon. The analysis was judged by the GDG
to be only partially applicable to this guideline review and the NICE reference case.
Also, it was based on a small pre- and post-observational study. Consequently, it
was judged by the GDG to have potentially serious methodological limitations.
12.3.6 Linking evidence to recommendations (non-acute community

mental healthcare)
The GDG agreed that the main aim of the EIS, CMHTs and ICM community-based
care is to provide evidence-based treatments in a community setting and thereby to
prevent or reduce admissions. However, each team or service-level intervention has
certain nuances in the aim and content of the intervention, and the patient
population they target, which influences which critical outcomes are relevant for
each team/service intervention. The GDG therefore decided on the following critical
outcomes.
EIS:
• Adverse events (for example, suicide)
• Service use
• Response/relapse
• Employment and education
• DUP
• Satisfaction with services (service user and carer)

CMHTs:
• Service use
• Employment and accommodation
• Quality of life
• Symptoms of psychosis and mental health
ICM:
• Loss to services
• Service use
• Quality of life
Early intervention services

EIS is a way of providing more intensive, personalised care for people in the first 3
years following first episode psychosis. From this review, EIS is better than
comparators (standard care or a CMHT) on a range of outcomes, including reduced
relapse rates, reduced hospital stay, improvement in symptoms and quality of life
and, importantly, EIS is preferred to standard services. EIS provided a range of
evidence-based interventions not routinely provided by other services (that is,
family intervention and CBT).
The review of psychological treatments for the 2009 guideline suggested that family
intervention for people with early psychosis reduces relapse rates but does little to
reduce symptoms, whereas CBT for psychosis reduced symptoms and improved
quality of life but did nothing to alter relapse rates. EIS teams included in the review
all provided family intervention and CBT. The GDG considered this and took the
view that although EIS providers often cite small caseloads and other factors, such as
team ethos, as the key ingredients leading to positive outcomes, the inclusion of
evidence-based psychological and pharmacological treatments in the context of such
small caseloads was probably a more likely explanation for the success of EIS.
Importantly, the review for this 2014 guideline included data not previously
available on the effects of EIS over 12 months after the end of treatment, which
suggests that the impact of EIS is lost by this stage. In practice, EIS currently
discharge people with early psychosis to CMHTs and other community services at
the end of 3 years. Therefore, to maintain benefits, service users should either remain
within EIS for longer periods or community teams (CMHT and ACT) for people
with established schizophrenia will need to provide the same evidence-based
treatments available in EIS, such as pharmacological, psychological and arts
therapies and support for employment provided within an integrated team.

Implications for all teams and services for people with psychosis and
schizophrenia
Following the review of EIS, the GDG considered the implications for all teams
providing services for psychosis and schizophrenia. EIS, more than any other service
developed to date, is associated with improvements in a broad range of critical
outcomes, including relapse rates, symptoms, quality of life and a better experience
for services. EISs reviewed here all included family interventions and CBT for
psychosis. The GDG took the view that, not only should EIS provide the full range of
evidence-based treatments recommended in this guideline, but all teams and
services should do so, irrespective of the orientation or type of team or service
considered. Thus, ICM teams, inpatient teams and CRHTTs should provide, or give
access to, pharmacological interventions, psychological interventions and any other
treatments recommended in this guideline. Moreover, EIS has a very modern
orientation to service user experience, which the GDG considered was encapsulated
by the existing NICE guidance and quality standard on Service User Experience in
Adult Mental Health (NICE, 2011) which covers community and hospital settings. The
GDG therefore decided to recommend that all teams providing care for people with
psychosis and schizophrenia should not only provide evidence-based treatments,
but they should also comply with Service User Experience in Adult Mental Health in the
way in which they deliver care.
Community mental health teams

The review for CMHTs included three trials, of which one was a cluster randomised
trial. The trial population was recruited from various sources, that is, those being
discharged from inpatient or outpatient treatment. Comparators were also mixed
and included participants receiving outpatient, inpatient and home treatment. Trials
included in the review were UK based (one in Manchester and two in London) but
were conducted in the 1990s. For people with severe mental illness, the GDG found
no evidence of a difference in effectiveness between CMHTs and standard care for
various symptom-related, service use and functioning outcomes. The most the GDG
could conclude from this is that in the mid-1990s CMHTs showed no superiority
over other ways of delivering care. The evidence is inconclusive and of historical
interest only.
Intensive case management

The dataset used for the review of ICM (24 trials of ICM, including ACT) was
relatively large compared with those used for other reviews of team and service-
level interventions. ICM was defined as a team-based approach using assertive case
management/care programming. In comparison with standard care, ICM was found
to be more effective than standard care for various critical outcomes including
reducing time spent in hospital, better engagement with services (from a proxy
measure of dropout from the trials), better quality of life and functioning, as well as
greater satisfaction with services. Furthermore, ICM was found to be equally as
effective as standard care for relapse rates and symptoms of psychosis, which
suggests that ICM is not harmful for people with psychosis and schizophrenia.
However, this benefit was not consistently found at longer follow-up points.

When compared with non-ICM (ICM defined as a caseload of 15 or less and non-
ICM as a caseload of more than 15), although no differences were observed in
symptoms, ICM was more effective at service user engagement at short-term follow-
up but this effect was not observed at longer follow-up points.
In a UK only sub-analysis most beneficial effects were no longer observed but ICM
was still beneficial for engagement and satisfaction with services compared with
standard care, which suggests that it is well tolerated and liked by service users. UK
data also suggest that ICM is no better than case management in the outcome of
interest. The lack of benefit of ICM could reflect the difficulty in reducing already
low bed numbers in the UK and that other outcomes, such as people’s views and
satisfaction with services, may be more appropriate to evaluate (Priebe et al., 2009).
The GDG also considered the qualitative data on the adaptation of ICM in the UK,
the CPA, which suggests service users do not value this approach and see it as
bureaucratic and defensive.
Trade-off between net health benefits and resource use:
Early intervention services

The UK-based economic evidence for EIS is based on two studies. One concluded
that EIS provides better outcome at no extra cost, and thus is cost effective at 18
months. Similarly, in the other UK study EIS was found to be cost saving over 3
years. The UK findings are supported by international evidence. However, weak
long-term clinical evidence associated with EIS means that there is uncertainty in the
results. Nevertheless, the GDG judged that the costs of providing such interventions
are justified by potential cost savings because of reduced relapse rates and shorter
hospital stays, and expected clinical benefits and improvements in the quality of life
of people with psychosis and schizophrenia.
Community mental health teams

The economic evidence for CMHTs is limited to one UK-based study. CMHTs
resulted in increased healthcare costs and poorer health outcomes compared with
EIS and consequently were not shown to be cost effective. Nevertheless, results
should be treated with caution since the difference in costs between interventions
was not significant and the clinical evidence pertaining to CMHTs is inconclusive.
Intensive case management

The economic evidence for ICM for individuals with psychosis and schizophrenia is
mixed. One UK study did not find any important clinical gains or cost savings. In
another UK study the costs of ICM were comparable to costs associated with
standard care and it resulted in greater levels of client satisfaction and engagement
with services. The international evidence on ICM is encouraging and although the
standard care in these studies is quite likely to be different from that in the UK, all of
the studies found ICM to be the preferred treatment strategy. Overall, the GDG

judged that the costs of providing ICM are justified by the expected savings arising
from shorter hospital stays and better engagement with the services.

The quality of the evidence base for these reviews ranged from very low to
moderate. Reasons for downgrading concerned risk of bias, high heterogeneity or
lack of precision in confidence intervals. Heterogeneity was a major concern when
evaluating the evidence. However, although variance was observed in the effect size
across studies, the direction of effect was consistent. Furthermore, sub-analysis for
UK-based studies resulted in more consistent findings, which suggests some
variance between UK-based and other studies in the content of both the active
intervention and the standard care comparator.
Overview of the evidence

The GDG took the view that the key to effectiveness for EIS is the provision of
evidence-based therapeutic interventions by competent practitioners. The GDG,
therefore, suggest that integrated, therapeutic community-based teams providing
evidence-based pharmacological, psychological and arts-based interventions, with
support for education and employment, consistent with other reviews in this
guideline, should be provided for people with psychosis and schizophrenia across
the age range. Particular care should be taken when engaging people with early
psychosis. The GDG felt that EIS or a specialist integrated community-based team
should initiate and continue treatment and care. The team should not have a focus
on risk-management but aim to engage the service user in services, and provide
support in an atmosphere of optimism and hope. The GDG also considered that
CMHTs represent an early stage in the evolution of community psychiatric care in
the UK and that the evidence suggests that team-based care is possible, not harmful.
The GDG considered the evidence for ICM and concluded that if engagement with,
and retention within, services is a clinical propriety, it appears to have some
advantages. Furthermore, the evidence suggests that smaller caseloads may not be
necessary, but this was likely to depend upon the severity of illness and level of
impairment of service users; finally the GDG judged that the CPA should be
replaced with a lower intensity, less bureaucratic and defensive case management
approach.

12.3.7.1 Use this guideline in conjunction with Service user experience in adult
mental health (NICE clinical guidance 136) to improve the experience of care
for people with psychosis or schizophrenia using mental health services,
and:
• work in partnership with people with schizophrenia and their
carers
• offer help, treatment and care in an atmosphere of hope and
optimism

• take time to build supportive and empathic relationships as an
essential part of care. [ 2009, amended 2014]
12.3.7.2 All teams providing services for people with psychosis or schizophrenia
should offer a comprehensive range of interventions consistent with this
guideline. [2009]
12.3.7.3 Early intervention in psychosis services should be accessible to all people
with a first episode or first presentation of psychosis, irrespective of the
person’s age or the duration of untreated psychosis. [new 2014]
12.3.7.4 People presenting to early intervention in psychosis services should be
assessed without delay. If the service cannot provide urgent intervention for
people in a crisis, refer the person to a crisis resolution and home treatment
team (with support from early intervention in psychosis services). Referral
may be from primary or secondary care (including other community
services) or a self- or carer-referral. [new 2014]
12.3.7.5 Continue treatment and care in early intervention in psychosis services or
refer the person to a specialist integrated community-based team. This team
should:
• offer the full range of psychological, pharmacological, social and
occupational interventions recommended in this guideline
• be competent to provide all interventions offered
• place emphasis on engagement rather than risk management
• provide treatment and care in the least restrictive and stigmatising
environment possible and in an atmosphere of hope and optimism
in line with Service user experience in adult mental health (NICE
clinical guidance 136). [new 2014]
12.3.7.6 Early intervention in psychosis services should aim to provide a full range of
pharmacological, psychological, social, occupational and educational
interventions for people with psychosis, consistent with this guideline.
[2014]
12.3.7.7 Consider extending the availability of early intervention in psychosis
services beyond 3 years if the person has not made a stable recovery from
psychosis or schizophrenia. [new 2014]
12.3.7.8 Consider intensive case management for people with psychosis or
schizophrenia who are likely to disengage from treatment or services. [new
2014]

12.3.8.1 How can the benefits of early intervention in psychosis services be
maintained once service users are discharged after 3 years? (see Appendix
10 for further details) [2014]

12.4 ALTERNATIVES TO ACUTE ADMISSION
12.4.1 Introduction
Home-based alternatives to acute admission
Diverting patients from admission has been one of the central purposes of
innovations in mental health service delivery for many decades; whereas it is only
relatively recently that preventing admission has become a focus of interest in the
rest of healthcare in the UK. The principal drivers for this have been the
unpopularity of overcrowded psychiatric wards, the involuntary aspects of mental
healthcare within hospitals and their high costs. Other arguments for home
treatment have been that patients’ autonomy and social functioning may be better
preserved when they are not admitted, that resolving the crisis at home may allow
skills for coping with future crises in the community to be enhanced, and
intervening with social triggers for crises and involving social networks is more
readily achieved (Johnson & Needle, 2008).
Innovative services assessing and treating service users at home in crises have been
established and evaluated in several countries since ArieQuerido first established a
programme to avert psychiatric admissions in Amsterdam in the 1930s (Hoult, 1991;
Johnson, 2013; Polak et al., 1979; Querido, 1935). Some of these services have been
free-standing crisis management teams, where patients were admitted at the time of
threatened admission to hospital and discharged once the crisis has resolved. Several
of the earlier innovative teams involving acute home treatment were hybrids of the
crisis team and ICM models, recruiting patients to home treatment at the time of a
crisis but then retaining them on caseloads in the longer term (Marks et al., 1994;
Stein & Test, 1980).
Community residential alternatives

Staying at home during a crisis is preferred by many service users, but not always
practical or desirable. The risk of harm to self or others is too great for some patients
to be left alone for extended periods of time without supervision. Others may be
severely functionally impaired, have no fixed abode, or live in environments that
exacerbate their difficulties. Residential alternatives outside hospital, such as crisis
houses, are a potential resource for people in crisis who cannot appropriately be
treated at home but who does not wish to go to hospital.
Residential crisis services in the community have a history spanning many decades,
but have not so far been implemented nationwide in any country. This is despite
strong advocacy by service user groups. Crisis houses are the most prevalent
community model: these are small unlocked, stand-alone community units that are
usually based in converted residential premises. An early innovative model was the
Soteria house in California in the early 1970s, subsequently emulated by services in a
several European countries (Bola & Mosher, 2002; Ciompi et al., 1995).

A comprehensive UK survey of alternatives to admission identified a variety of
models, from services that followed a largely clinical model, with mental health
professional staff and types of care similar to those on acute wards, to more radical
alternatives aiming to provide treatment approaches significantly different from
hospitals, often managed by third sector organisations (Johnson et al., 2009). Most of
the alternatives found worked closely with CRHTTs and were well integrated into
catchment area mental health systems. Family sponsor homes, where people in crisis
are hosted by carefully selected and trained families, usually also with the support of
the CRHTT, are another community model for avoiding admission (Aagaard et al.,
2008), although few such schemes are currently available in the UK.
Ethical and practical difficulties in recruiting patients to trials at the time of a crisis,
and resistance to randomisation in well-established often third sector-provided
alternatives, have recently limited the conduct of RCTs of crisis houses and other
residential alternatives. However, a small number of trials, generally with
populations too diagnostically mixed to be within the scope of this guideline, have
tended to report better patient satisfaction and otherwise similar outcomes for crisis
houses compared with inpatient wards (Howard, 2010; Lloyd-Evans et al., 2009).
Implementation studies of the model have suggested that service user populations
are similar to hospital wards, but with most patients voluntary and already known
to services and with significantly less risk of violence than hospital patients (Johnson
et al., 2009). Naturalistic investigation using quantitative and qualitative methods
has also indicated a marked service user preference for crisis houses rather than
wards, supporting strong voluntary sector advocacy for these services (Gilburt et al.,
2010; Mind, 2011; Osborn et al., 2010b). An investigation of the views of local
stakeholders, including referrers and senior managers, suggested that acute
residential services in the community were valued as a means of extending service
user choice and available strategies for managing crises. They were also seen as
taking pressure off hard-pressed hospital inpatient services by means that included
diverting patients who would otherwise have been admitted, accepting early
discharges and providing respite to people at potentially high risk of reaching the
admission threshold without additional support (Morant et al., 2012).
A recent trend in development of crisis residential alternatives has been towards

close integration between crisis teams and crisis houses - the ability of each to
manage challenging patients in the community might potentially be enhanced
through synergy with the other.
12.4.2 Crisis resolution and home treatment teams

Introduction
England is one of very few countries in which provision of acute home treatment
services has been national policy, with all trusts required to introduce CRHTTs (also
known in some areas as crisis assessment and treatment teams or intensive home
treatment teams) under the NHS Plan (Department of Health, 2000). While provision
of such services is no longer mandatory, they remain very widespread in the UK.

The primary aims of CRHTTs are to:
• assess all patients being considered for admission to acute psychiatric wards
• initiate a programme of home treatment with frequent visits (usually at least
daily) for all patients for whom this appears a feasible alternative to hospital
treatment
• continue home treatment until the crisis has resolved and then transfer people
to other services for any further care they may need
• facilitate early discharge from acute wards by transferring inpatients to
intensive home treatment.
The teams are multidisciplinary, usually containing nurses, psychiatrists and non-
professional mental health staff such as support workers, with occupational
therapists, psychologists, social workers and clinical psychologists less consistently
represented. Guidance on model implementation suggests they should operate 24
hours a day 7 days a week, and most at least work extended hours. Gatekeeping
acute beds, with no hospital admissions taking place unless the CRHTT confirms
that home treatment does not appear feasible, is regarded as a key activity associated
with success in reducing acute bed use (Middleton et al., 2008). Accounts of the
model suggest that core team interventions should include: visiting at home (at least
twice a day if needed) to provide support and monitor recovery from the crisis and
risk; prescribing, dispensing and monitoring adherence to medication; helping
resolve practical problems that may perpetuate the crisis; brief psychological and
social interventions to alleviate symptoms and distress and reinforce coping skills
and problem solving abilities; and support for carers and other key social network
members (Johnson, 2013). The team’s work is short term, with discharge to any
services required for long-term support generally taking place within a few weeks.

A Cochrane review of crisis interventions for people with serious mental health
problems (Murphy et al., 2012) was identified and selected by the GDG for review
and further analysis.
The GDG adopted the inclusion criteria and definition of crisis resolution developed
by the Cochrane review for studies of CRHTTs in the management of people with
severe mental illness. Crisis intervention and the comparator treatment were defined
as follows:
• crisis resolution is any type of crisis-orientated treatment of an acute
psychiatric episode by staff with a specific remit to deal with such situations,
in and beyond ‘office hours’
• ‘standard care’ is the normal care given to those experiencing acute
psychiatric episodes in the area concerned; this involved hospital-based
treatment for all studies included.

The focus of the review was to examine the effects of CRHTT care for people with
severe mental illness experiencing an acute episode, compared with the standard
care they would normally receive.
Clinical review protocol (crisis resolution and home treatment teams)

The review protocol, including the review questions, information about the
databases searched, and the eligibility criteria used for this section of the guideline,
can be found in Table 142 (the full review protocol and a complete list of review
questions can be found in Appendix 6; information about the search strategy can be
found in Appendix 13).
Table 142: Clinical review protocol for the review of crisis resolution and home
treatment teams
potential harms of crisis resolution and home treatment teams compared with
treatment as usual or another intervention?
Objectives To evaluate the clinical effectiveness of crisis resolution and home treatment
teams in the treatment of psychosis and schizophrenia.
Intervention(s) Crisis resolution and home treatment teams
o Admission/readmission to hospital
o Number of days in hospital
o Number of staff/user contacts
• Satisfaction
• Mental health act use
Process
Study design RCTs
Sub-analysis

studies.

Six RCTs (N = 851) met the eligibility criteria for this review: FENTON1979 (Fenton
et al., 1979), HOULT1983 (Hoult et al., 1983), JOHNSON2005 (Johnson et al., 2005),
MUIJEN1992 (Muijen et al., 1992), PASAMANICK1964 (Pasamanick et al., 1964),
STEIN1975 (Stein et al., 1975). All six were published in peer-reviewed journals
between 1964 and 2005, and all compared CRHTTs with standard care as defined by
the study. The Cochrane review of crisis interventions (Murphy et al., 2012) was
used as a source to verify that all relevant studies had been included. Further
15a.Table 143 provides an overview of the included trials.
Table 143: Study information table for trials included in the meta-analysis of
CRHTTs versus standard care
CRHTTs versus standard care

Study ID(s) FENTON1979
HOULT1983
JOHNSON2005
MUIJEN1992
PASAMANICK1964
STEIN1975
Country Australia (k = 1)
Canada (k = 1)
UK (k = 2)
US (k = 2)
Mean age of participants (range) 35.76 years (30.95 to 40.08 years)1
Mean percentage of participants with primary 74.29% (53 to 100%)2
Mean gender % women (range) 53.14% (41.38 to 68%)
Intervention type Community Living Program’s home-based care (k = 1)
Daily Living Program’s home-based care (k = 1)
Home crisis care by CRHTTs (k = 1)
Home Care Group (k = 3)
Comparisons Standard care: hospitalisation (k = 5)
Standard care from the inpatient unit, crisis houses, and
CMHTs (k = 1)
Note.1FENTON1979 and HOULT1983 did not provide data.
2 STEIN1975did not provide data.
Clinical evidence for crisis resolution and home treatment teams


Evidence suggest that CRHTTs, when compared with standard care, reduce the
likelihood of people with serious mental health problems being admitted to
inpatient settings at up to 6 months (k = 3; N = 325), 12 months (k = 3; N = 400) and
at 24 months’ follow-up (k = 1; N = 118). However, the evidence was of either very
low or low quality. Nevertheless, the size of the effects in reducing admission at each
time interval was large.
However, very low quality evidence showed that CRHTTs were no more effective
than standard care in reducing the likelihood of people with serious mental illness
being readmitted at either 12 months’ (k = 4; N = 601) or 24 months’ follow-up (k = 2;
N = 306). The evidence in this area is inconclusive.
Table 144: Summary of findings tables for CRHTTs compared with standard care

Intervention: CRHTTs
Outcomes Illustrative comparative risks* (95% CI) Relative No. of Quality of the
Assume Corresponding risk effect participants evidence
d risk (95% CI) (studies) (GRADE)
TAU CRHTTs
Service use Study population RR 0.35 205 ⊕⊝⊝⊝
(admitted to 854 per 299 per 1000 (0.11 to (2 studies) Very low1,2,3
hospital) - by 3 1.18)
1000 (94 to 1000)
months
833 per 292 per 1000
1000 (92 to 983)
(admitted to 904 per 253 per 1000 (0.09 to (3 studies) Very low1,2,3
hospital)- by 6 1000 (81 to 795) 0.88)
months
900 per 252 per 1000
1000 (81 to 792)
Service use Study population RR 0.4 400 ⊕⊕⊝⊝
(admitted to 990 per 396 per 1000 (0.31 to (3 studies) Low1,4
hospital) - by 12 1000 (307 to 505) 0.51)
months
1000 per 400 per 1000
1000 (310 to 510)
Service use Study population RR 0.32 118 ⊕⊕⊝⊝
(admitted to 1000 per 320 per 1000 (0.22 to (1 study) Low5,6
hospital) - by 24 1000 (220 to 460) 0.46)
months
1000 per 320 per 1000
1000 (220 to 460)
(readmitted to 402 per 205 per 1000 (0.21 to (4 studies) Very low1,2,3
hospital) - by 12 1000 (84 to 482) 1.2)
months
451 per 230 per 1000
1000 (95 to 541)
(readmitted to (0.36 to (2 studies) Very low1,2,3
391 per 297 per 1000

hospital) - by 24 1000 (141 to 637) 1.63)
months 407 per 309 per 1000
1000 (147 to 663)
Mental Health Act Study population RR 0.65 87 ⊕⊕⊝⊝
admission - by 3 310 per 201 per 1000 (0.31 to (1 study) Low3,5
months 1000 (96 to 418) 1.35)
310 per 201 per 1000
1000 (96 to 419)
Satisfaction (patient N/A Mean satisfaction (patient N/A 115 ⊕⊕⊝⊝
satisfied with care - satisfied with care - Satisfaction (1 study) Low5,6
Satisfaction Scale) - Scale - by 6 months) in the
by 6 months intervention groups was
(0.57 to 1.34 higher)
Satisfaction Scale) - Scale - by 12 months) in the
(0.64 to 1.4 higher)
Satisfaction Scale) - scale - by 20 months) in the
(0.85 to 1.58 higher)
Satisfaction (patient Study population RR 1.04 87 ⊕⊕⊝⊝
not satisfied with 405 per 421 per 1000 (0.63 to (1 study) Low3,5
care - CSQ) - by 3 1000 (255 to 696) 1.72)
months
286 per 297 per 1000
1000 (180 to 492)
Note. CI: Confidence interval; RR: Risk ratio; TAU = treatment as usual; CSQ =
3CI crosses the clinical decision threshold (SMD of 0.2 or -0.2; RR of 0.75 or 1.75).
estimate of effect.
6 Criteria for an optimal information size not met
Low quality evidence from a single study (N = 87) reported no difference in rate of
Mental Health Act admission or in satisfaction with care between CRHTT and
standard care at 3 months’ follow-up. However, at 6 (k = 1; N = 115), 12 (k = 1; N =
121) and 20 months’ follow-up (k = 1; N = 137) low quality evidence showed that
those who received care from a CRHTT reported greater satisfaction with care
compared with those who received standard care.

It was decided by the GDG not to use the data available on the duration of acute
inpatient care. This was because four studies included ‘index admission’ in their data
and were therefore deemed unrepresentative.

For people with schizophrenia and other serious mental health problems in an acute
crisis, care from a CRHTT is superior to standard hospital care in reducing hospital
admissions and appears to be more acceptable at long-term follow-up. CRHTTs also
appear to increase retention of service users, improve quality of life and have a
marginally better effect on some clinical outcomes.

The systematic literature search identified two UK-based economic studies that
assessed the economic impact of CRHTTs for individuals with psychosis and
schizophrenia (McCrone et al., 2009a; McCrone et al., 2009b). Details on the methods
used for the systematic search of the economic literature are described in Chapter 3.
References to included studies and evidence tables for all economic studies included
in the guideline systematic literature review are presented in Appendix 19.
Completed methodology checklists of the studies are provided in Appendix 18.
Economic evidence profiles of studies considered during guideline development
(that is, studies that fully or partly met the applicability and quality criteria) are
presented in Appendix 17, accompanying the respective GRADE clinical evidence
profiles.
McCrone and colleagues (2009a) conducted a cost-effectiveness analysis that

compared CRHTTs with standard care. Standard care was defined as care by
CMHTs, inpatient care and crisis houses. Study population comprised service users
with psychosis, schizophrenia, bipolar affective disorder, unipolar depression,
personality disorder, and non-psychotic disorder (<5%). The study was based on a
large RCT (JOHNSON2005) (n = 260) and a public sector payer perspective was
adopted. The time frame of the analysis was 6 months. The authors considered NHS
costs (primary, secondary, and community care) and criminal justice sector costs
incurred by prison and police cell stay. The primary outcome was the number of
days not on a psychiatric ward or other inpatient setting. Costs were reported
including and excluding inpatient care. Costs per person inclusive of inpatient care
were lower in the CRHTTs group by £2,438 (p < 0.01) in 2003/04 prices, however if
inpatient care was excluded the costs per person were higher by £768 (p < 0.01) in
the CRHTT group. Days not on psychiatric ward per service user were very similar
in both groups: 126.8 versus 129.9 days for CRHTTs and standard care groups,
respectively. Cost-effectiveness analysis, excluding inpatient costs, showed that if
society is willing to pay £100 to avoid an extra inpatient day, the probability of
CRHTTs being cost effective would be 1.00. Even though the analysis has included
criminal justice sector costs these costs accounted for only a very small proportion of
the total costs and thus are unlikely to affect the results. Also, the authors made no
attempt to estimate QALYs, however this did not affect judgement on cost
effectiveness since clinical outcomes were very similar. Consequently, the analysis

was judged by the GDG to be directly applicable to this guideline review and the
NICE reference case. The time horizon of the study was only 6months, which may
not be sufficiently long enough to fully capture the effects of the intervention.
However, taking into account data limitations, overall the analysis was judged by
the GDG to have only minor methodological limitations.
Another cost analysis by McCrone and colleagues (2009b) compared CRHTTs with
standard care. Standard care included care in acute wards, crisis houses, care by
CMHTs and liaison teams based in the local casualty department. The study was
based on a pre- and post-observational study (n = 200) that mainly included
individuals with schizophrenia/schizoaffective disorder and bipolar affective
disorder. The study adopted a public sector payer perspective and considered costs
over a 6-month period. The analysis included NHS costs (inpatient, outpatient and
community care) and also criminal justice sector costs incurred by arrest, solicitor,
court appearance, police, probation, and police cell/prison. The authors adjusted
costs for the baseline differences in participant characteristics and estimated that
CRHTTs resulted in cost savings of £1,681 (p = ns) in 2001 prices. The sensitivity
analysis showed that if the unit cost of contact with the CRHTT was £40, the cost
difference would increase to -£1,807 (p < 0.1). Also, if groups were defined according
to whether there was any CRHTT contact, the cost savings would increase to £2,189
(p < 0.1). The analysis was only partially applicable to this guideline review since it
included costs accruing to the criminal justice sector. Healthcare and crime costs
were not reported separately; consequently it is not clear what proportion of the
total costs are accounted for by contacts with the criminal justice system. The
analysis was based on a pre- and post-observational study, which are prone to bias
because of the inability to control for confounding factors. However, the authors
used a regression approach to control for a range of confounders. As a result this
study was judged by the GDG to have only minor methodological limitations.
12.4.3 Crisis houses

Introduction
Crisis houses are a residential alternative to acute care in a crisis that can be
provided to support the care provided by the local CRHTT. They are designed to be
a ‘home away from home’ based in the local community for people who are
experiencing a crisis. Crisis houses are staffed 24 hours a day either by trained
mental health staff and based within mental health services, or by support workers
trained in crisis care and based within voluntary sector organisations. In the latter
context, crisis house workers are usually supported by the local CRHTT.
The service user’s treatment and medication management is sometimes the

responsibility of the mental health team running the crisis house; sometimes their
community-based psychiatrist and sometimes by the CRHTT. Usually, however,
workers in the crisis house assist with treatment planning and offer day-to-day
support for community-based treatment, employment or education, or other
community-based social activities that can help the service user’s social functioning

and activities of daily living. They also sometimes offer transportation to and from
treatment facilities and community or outpatient appointments. The service user
sleeps at the crisis house overnight with trained support workers or trained mental
health staff available 24 hours a day.

A crisis house is defined as a residential alternative to acute admission during a
crisis. A crisis house aims to help the service user maintain autonomy and normality
during a crisis within their own community but is also supported with their
treatment plan and daily living, allowing an easier transition back to normal life
after the crisis. Crisis houses also aims to reduce the stigma of experiencing a crisis,
which sometimes may be exacerbated by admission to an inpatient facility, allowing
the service user and families to move away from the idea of the service user being
‘unwell’ and providing the support needed for swift recovery.
Clinical review protocol (crisis houses)

questions can be found in Appendix 6; further information about the search strategy
can be found in Appendix 13).
Table 145: Clinical review protocol for the review of crisis houses
potential harms of crisis houses compared withtreatment as usual or another
intervention?
Objectives To evaluate the clinical effectiveness of crisis houses in the treatment of
psychosis and schizophrenia.
Intervention(s) Crisis houses
o Admission/ Readmission to hospital
o Number of days in hospital
o Number of staff/user contacts
• Satisfaction
• Mental Health Act use
Process
Date searched SR/RCT: Inception to June 2013
Study design RCTs

Sub-analysis

studies.
One RCT (N = 185) providing relevant clinical evidence met the eligibility criteria for
this review. The study was published in a peer-reviewed journal in 1998. Further
15a.
The one study compared crisis houses with standard care. Table 146 provides an
overview of the included trial.
Table 146: Study information table for trials included in the meta-analysis of crisis
houses versus standard care
Crisis houses versus standard care

Study ID FENTON1998
Country USA
Year of publication 1998
Mean age of participants 37.58 years
Mean percentage of participants with primary diagnosis of 56%
psychosis of schizophrenia
Mean gender % women 47.9%
Length of follow-up 26 weeks
Intervention type Home-like acute residential facility (k = 1)
Comparisons Standard care (k = 1)
Clinical evidence for crisis houses

presented in Table 147.

Table 147: Summary of findings tables for crisis houses compared with standard
care

Intervention: Crisis houses
TAU Crisis houses (recovery houses) (GRADE)
Service use RR 1 185 ⊕⊕⊝⊝
Study population
(admitted to (0.98 to (1 study) Low1
hospital) - by 6 1000 per 1000 per 1000 1.02)
months follow-up 1000 (980 to 1000)
1000 per 1000 per 1000
1000 (980 to 1000)
Service use RR 0.9 185 ⊕⊕⊝⊝
Study population
(readmitted to (0.76 to (1 study) Low2,3
hospital) - by 6 804 per 724 per 1000 1.05)
months follow-up 1000 (611 to 845)
804 per 724 per 1000
1000 (611 to 844)
Service use (days of N/A Mean service use (days of acute N/A 108 ⊕⊕⊝⊝
acute inpatient care) inpatient care - by 6 months (1 study) Low2,3
- by 6 months follow-up) in the intervention
follow-up groups was 0.02 standard
0.36 higher)
Service use (number N/A Mean service use (number of N/A 111 ⊕⊕⊝⊝
of repeat admissions repeat admissions per participant (1 study) Low2,3
per participant) - by - by 6 months follow-up) in the
6 months follow-up intervention groups was
Note. CI: Confidence interval; RR: Risk ratio; TAU = treatment as usual
*The basis for the assumed risk (for example, the median control group risk across studies) is provided in
the footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the
1 Criteria for an optimal information size not met.

3CI crosses the clinical decision threshold (SMD of 0.2 or -0.2; RR of 0.75 or 1.75).
Low quality evidence showed no additional benefit of crisis houses, when compared
with standard care, on hospital admission (k = 1; N = 185), hospital readmission (k =
1; N = 185), number of days spent in acute care (k = 1; N = 108) or the number of
repeat admissions per participant (k = 1; N = 111) at 6 months’ follow-up. No data
were available on satisfaction or Mental Health Act admissions. The data were
considered by the GDG to be inconclusive.

The data available from a single study was inconclusive.

No studies assessing the cost effectiveness of crisis houses for adults with psychosis
or schizophrenia were identified by the systematic search of the economic literature
undertaken for this guideline. Details on the methods used for the systematic search
of the economic literature are described in Chapter 3.
12.4.4 Acute day hospital care

Introduction
Given the substantial costs and high level of use of inpatient care, the possibility of
day hospital treatment programmes acting as an alternative to acute admission
gained credence in the early 1960s, initially in the US (Kris, 1965; Herz et al., 1971),
and later in Europe (Wiersma et al., 1989) and the UK (Creed et al., 1990; Dick et al.,
1985). Acute day can be provided to support the care provided by the local CRHTT.

A Cochrane review of acute day hospitals for people with serious mental health
problems (Marshall et al., 2011) was identified and selected by the GDG for review
and further analysis.
The GDG adopted the inclusion criteria and definition of acute day hospitals
developed by the Cochrane review. Acute day hospitals and the comparator
treatment were defined as follows:
• Acute day hospitals were defined as units that provided ‘diagnostic and
treatment services for acutely ill individuals who would otherwise be treated
in traditional psychiatric inpatient units’ (Rosie, 1987).
• Standard care was defined as admission to an inpatient unit.
Thus, trials would only be eligible for inclusion if they compared admission to an
acute day hospital with admission to an inpatient unit. Participants were people
with acute psychiatric disorders (all diagnoses) who would have been admitted to
inpatient care had the acute day hospital not been available.
Clinical review protocol (acute day hospitals)

questions can be found in Appendix 6; further information about the search strategy
can be found in Appendix 13).

Table 148: Clinical review protocol for the review of acute day hospital treatment
potential harms of acute day hosiptals compared with standard care?
Objectives To evaluate the clinical effectiveness of acuetd ay hospitals in the treatment of
Intervention(s) Acute day hospitals
Comparison Standard care
emergency services)
• Satisfaction
o User satisfaction (validated measures only)
o Carer satisfaction (validated measures only)
Process
Study design RCTs
Sub-analysis
Where data were available, sub-analyses were conducted for UK only studies.
The GDG selected an existing Cochrane review (Marshall et al., 2011) as the basis for
this section of the guideline, with a new search conducted to update it. This
Cochrane review is an update of the previous Health Technology Appraisal
(Marshall et al., 2001) of nine trials with the addition of a large EU multi-centre trial
(Kallert-EU-2007). A search for recent RCTs did not uncover any suitable new
studies to add to the Marshall review. The existing Cochrane review included ten

RCTs (N = 2685) providing relevant clinical evidence meeting the eligibility criteria
for the review. Studies were published in peer-reviewed journals between 1965 and
2007. Further information about included studies can be found in Appendix 15a.
Further information about excluded studies can be found in (Marshall et al., 2011)
Of the ten included trials, all compared acute day hospitals with routine inpatient
care. Table 149 provides an overview of the included trials.
Some difficulties were encountered in synthesising the outcome data because of the:
• Population
o Mixed sample both within and between studies and only a quarter to a
third had a diagnosis of schizophrenia in the included studies
o Day hospital care was unsuitable for some people and a proportion of
studies excluded these people prior to randomisation
o Country
 The setting of trials varied across studies. EU multicentre (k = 1);
US (k = 4); Netherlands (k = 2); UK (k = 3).
• Intervention
o Some interventions included additional services (for example, out-of-
hours back-up, ‘back-up bed’) while others did not.
• Methods
o The point of randomisation varied across studies (unsuitable patients
excluded prior to randomisation or randomisation at referral).
• Outcomes
o A number of similar outcomes were presented in slightly different
formats across studies.
• Follow-up
o Follow-up varied from 2 to 24 months between studies.
Table 149: Study information table for trials included in the meta-analysis of acute
day hospital versus standard care
Acute day hospital treatment teams versus

standard care
Study ID(s) Creed-UK-1990
Creed-UK-1996
Dick-UK-1985
Herz-US-1971
Kallert-EU-2007
Kris-US-1965
Schene-NL-1993
Sledge-US-1996
Wiersma-NL-1989
Zwerling-US-1964
Country Europe (k = 1)
Netherlands (k = 2)
UK (k = 3)
US (k = 4)

Mean age of participants (range) 37.2 years (32 to 42.38 years)1
Mean percentage of participants with primary 32.68% (23.5 to 39%)2
Intervention type Acute day hospital treatment (k = 10)
Comparisons Routine inpatient care (k = 10)
Note.1Dick-UK-1985, Kris-US-1965, Schene-NL-1993 did not provide data.
2Dick-UK-1985, Kris-US-1965, Schene-NL-1993, Zwerling-US-1964 did not provide data.
Clinical evidence for acute day treatment

presented
Intervention: Acute day hospitals
Comparison: Inpatient admission
Assumed Corresponding risk effect participants the evidence
Inpatient Acute day hospitals
admission
Feasibility and Study population RR 0.97 1,117 ⊕⊕⊕⊕
engagement -lost 282 per 274 per 1000 (0.80 to (1 study) High
to follow-up - end 1000 (226 to 330) 1.17)
of study (by 3
months) - type 1
studies
Feasibility and Study population RR 0.83 312 ⊕⊕⊝⊝
engagement - lost 315 per 262 per 1000 (0.58 to (2 studies) Low1,3
of study (by 2-6
months) - type 1
studies
Feasibility and Study population RR 0.94 1,704 ⊕⊕⊕⊝
engagement - lost 327 per 307 per 1000 (0.82 to (5 studies1) Moderate2
of study (by 1
year) - type 1
studies
Duration of index N/A Mean duration of index admission N/A 1,582 ⊕⊕⊕⊝
admission (days/ (days/month) in the intervention (4 studies1) Moderate2
month) - type 1 groups was 27.47 higher (3.96 to
studies 50.98 higher)
Duration of all N/A Mean duration of all hospital care N/A 465 ⊕⊕⊝⊝
hospital care (days/month) in the intervention (3 studies) Low3,4
(days/month) - groups was 0.38 lower (1.32 lower
type 1 studies to 0.55 higher)
Duration of stay in N/A Mean duration of stay in hospital N/A 465 ⊕⊕⊝⊝
hospital (days/month) in the intervention (3 studies) Low3,4
(days/month) - groups was 2.75 lower (3.63 to 1.87
type 1 studies lower)

Duration of all day N/A Mean duration of all day patient N/A 465 ⊕⊕⊝⊝
patient care care (days/month) in the (3 studies) Low2,3
(days/month) - intervention groups was 2.34
type 1 studies higher (1.97 to 2.70 higher)
Readmitted to day/ Study population Not 667 ⊕⊕⊝⊝
inpatient care after estimable (5 studies) Low3,4
311 per 0 per 1000
discharge (days/
1000 (0 to 0)
month)- type 1
studies
Satisfaction with Study population RR 0.46 91 ⊕⊕⊕⊝
services - not (0.27 to (1 study) Moderate3,4
604 per 278 per 1000
satisfied with care 0.79)
1000 (163 to 477)
received - type 1
studies
engagement - lost (0.48 to (1 study) Low3,4
509 per 351 per 1000
to follow-up (at 2 0.99)
years) - type 2 1000 (244 to 504)
studies
hospital care (days/ (days/months, individual patient (1 study) Low3,4
months, data – ‘nights in’ and ‘nights out’)
individual patient in the intervention groups was 1.10
data – ‘nights higher (1.58 lower to 3.78 higher)
in’and‘nights out’)
- type 2 studies
Readmitted to day/ Study population RR 0.93 160 ⊕⊕⊝⊝
inpatient care after (0.64 to (1 study) Low3,4
439 per 408 per 1000
discharge (days / 1.35)
1000 (281 to 592)
month) - type 2
studies
in the footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the
1 One large (n = 1,117) high-quality multi-centre RCT (Kallert-EU-2007) provides data for all outcomes.
This trial carries more weight than other pooled trials and this was taken into consideration when
assessing overall risk of bias.
2 Heterogeneity not explained by differences in populations/interventions.
3 Studies included are at a moderate risk of bias.
Clinical evidence for type 1 trials

Low to high quality evidence from up to five trials (N = 1,714) showed that there
was no difference between acute day hospitals and standard inpatient care in the
number lost to follow-up at the end of the intervention (between 3 months and 1
year). Kallert-EU-2007 also did not observe a significant difference between groups
in the number of participants lost to follow-up.
Moderate quality evidence from eight trials (N = 1,582) showed that participants in
the day hospital care group had significantly longer index admission than those in
the standard care inpatient group. This finding was mirrored by the Kallert-EU-2007

trial which found the duration of index admission was significantly longer in day
hospital settings than in standard inpatient care: 78 (SD = 73) versus 46 (SD = 46)
days (p<.001).
Low quality evidence from up to three trials (N = 465) showed no difference in all
hospital care between acute day hospitals and standard inpatient care. However, the
day patient group spent significantly longer in day patient care and significantly less
time in inpatient care than the standard care group.
Low quality evidence from up to five trials (N = 667) showed no difference between
day hospital care and standard inpatient care in the number of participants
readmitted to day/inpatient care after discharge.
One trial with 91 participants provided moderate quality evidence that day hospital
care was significantly more satisfactory than standard inpatient care. However, the
Kallert-EU-2007 trial provided no evidence of a difference between groups in
satisfaction with services (using a continuous measure).

One study with 160 participants provided low quality evidence favouring day
hospital care in the number of participants lost to follow-up. Low quality evidence
from one study (N = 160) showed no difference between groups in duration of all
hospital care or in the number of participants readmitted to day/inpatient care after
discharge.
Trials were categorised according the method of randomising participants. Marshall

and colleagues (2011) termed trials as type 1 and type 2. Type 1 trials were those in
which anyone considered ineligible for day hospital treatment was excluded before
randomisation (Creed-UK-1990, Creed-UK-1996, Dick-UK-1985, Herz-US-1971,
Kallert-EU-2007, Kris-US-1965, Schene-NL-1993, Sledge-US-1996.). In Type 2 trials,
everyone considered for admission to the acute day hospital service was
randomised, regardless of suitability; but anyone allocated to the acute day hospital
but who was too unwell for day hospital care was then admitted to the inpatient
ward (Wiersma-NL-1989 and Zwerling-US-1964.). Due to the methodological
differences, type 1 and type 2 trials are analysed separately.
In addition, the GDG decided that the large Kallert-EU-2007 trial provides a more
accurate depiction of service provision in the UK and increased confidence in the
findings of the review. Therefore, the GDG decided that the findings of this trial
should be assessed both as part of the meta-analysis and described individually to
assess if the findings are concurrent with the overall meta-analysis. Therefore,
relevant outcome findings from this trial are described narratively below.

Table 150: Summary of findings tables for acute day hospitals compared with
standard care

Intervention: Acute day hospitals
Comparison: Inpatient admission
Assumed Corresponding risk effect participants the evidence
Inpatient Acute day hospitals
admission
Feasibility and Study population RR 0.97 1,117 ⊕⊕⊕⊕
engagement -lost 282 per 274 per 1000 (0.80 to (1 study) High
of study (by 3
months) - type 1
studies
engagement - lost 315 per 262 per 1000 (0.58 to (2 studies) Low1,3
of study (by 2-6
months) - type 1
studies
Feasibility and Study population RR 0.94 1,704 ⊕⊕⊕⊝
engagement - lost 327 per 307 per 1000 (0.82 to (5 studies1) Moderate2
of study (by 1
year) - type 1
studies
Duration of index N/A Mean duration of index admission N/A 1,582 ⊕⊕⊕⊝
admission (days/ (days/month) in the intervention (4 studies1) Moderate2
month) - type 1 groups was 27.47 higher (3.96 to
studies 50.98 higher)
hospital care (days/month) in the intervention (3 studies) Low3,4
(days/month) - groups was 0.38 lower (1.32 lower
type 1 studies to 0.55 higher)
Duration of stay N/A Mean duration of stay in hospital N/A 465 ⊕⊕⊝⊝
in hospital (days/month) in the intervention (3 studies) Low3,4
(days/month) - groups was 2.75 lower (3.63 to 1.87
type 1 studies lower)
Duration of all day N/A Mean duration of all day patient N/A 465 ⊕⊕⊝⊝
patient care care (days/month) in the (3 studies) Low2,3
(days/month) - intervention groups was 2.34
type 1 studies higher (1.97 to 2.70 higher)
Readmitted to day/ Study population Not 667 ⊕⊕⊝⊝
inpatient care after 311 per estimable (5 studies) Low3,4
0 per 1000
discharge (days/ 1000 (0 to 0)
month)- type 1
studies
Satisfaction with Study population RR 0.46 91 ⊕⊕⊕⊝
services - not 604 per 278 per 1000 (0.27 to (1 study) Moderate3,4
satisfied with care 0.79)

received - type 1 1000 (163 to 477)
studies
engagement - lost 509 per 351 per 1000 (0.48 to (1 study) Low3,4
to follow-up (at 2 1000 (244 to 504) 0.99)
years) - type 2
studies
hospital care (days/months, individual patient (1 study) Low3,4
(days/ months, data – ‘nights in’ and ‘nights out’)
individual patient in the intervention groups was 1.10
data – ‘nights higher (1.58 lower to 3.78 higher)
in’and‘nights out’)
- type 2 studies
Readmitted to day/ Study population RR 0.93 160 ⊕⊕⊝⊝
inpatient care after 439 per 408 per 1000 (0.64 to (1 study) Low3,4
discharge (days / 1000 (281 to 592) 1.35)
month) - type 2
studies
in the footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the
1 One large (n = 1,117) high-quality multi-centre RCT (Kallert-EU-2007) provides data for all outcomes.
This trial carries more weight than other pooled trials and this was taken into consideration when
assessing overall risk of bias.
2 Heterogeneity not explained by differences in populations/interventions.
3 Studies included are at a moderate risk of bias.

Low to high quality evidence from up to five trials (N = 1,714) showed that there
was no difference between acute day hospitals and standard inpatient care in the
number lost to follow-up at the end of the intervention (between 3 months and 1
year). Kallert-EU-2007 also did not observe a significant difference between groups
in the number of participants lost to follow-up.
Moderate quality evidence from eight trials (N = 1,582) showed that participants in
the day hospital care group had significantly longer index admission than those in
the standard care inpatient group. This finding was mirrored by the Kallert-EU-2007
trial which found the duration of index admission was significantly longer in day
hospital settings than in standard inpatient care: 78 (SD = 73) versus 46 (SD = 46)
days (p<.001).
Low quality evidence from up to three trials (N = 465) showed no difference in all
hospital care between acute day hospitals and standard inpatient care. However, the
day patient group spent significantly longer in day patient care and significantly less
time in inpatient care than the standard care group.

Low quality evidence from up to five trials (N = 667) showed no difference between
day hospital care and standard inpatient care in the number of participants
readmitted to day/inpatient care after discharge.
One trial with 91 participants provided moderate quality evidence that day hospital
care was significantly more satisfactory than standard inpatient care. However, the
Kallert-EU-2007 trial provided no evidence of a difference between groups in
satisfaction with services (using a continuous measure).

One study with 160 participants provided low quality evidence favouring day
hospital care in the number of participants lost to follow-up. Low quality evidence
from one study (N = 160) showed no difference between groups in duration of all
hospital care or in the number of participants readmitted to day/inpatient care after
discharge.

There is no evidence of a difference between day hospital care and standard
inpatient care in engagement of participants. There is some evidence that the
duration of index admission is longer for participants in day hospital care. Although
no difference was observed between groups in the total days in hospital (day- or
inpatient), while the duration of day patient care is longer, the duration of inpatient
care is shorter for those in day hospital care. Although significantly more people
receiving day hospital care were satisfied with services, this difference was not
observed in the Kallert-EU-2007 trial.

No studies assessing the cost effectiveness of acute day hospitals for adults with
psychosis and schizophrenia were identified by the systematic search of the
economic literature undertaken for this guideline. Details on the methods used for
the systematic search of the economic literature are described in Chapter 3.
Given the large direct medical costs associated with relapse in psychosis and
schizophrenia, primarily resulting from expensive inpatient treatment, it has been
suggested that the lower operational cost of acute day hospitals could result in
substantial savings for the health service. On the other hand, there have been fears
that these savings would be achieved by shifting the cost burden to families and
carers, offering no real reduction in the overall cost to society. Nevertheless, the unit
cost of acute inpatient care per bed day is £330 in 2011/12 prices (Curtis, 2012). This
estimate has been based on the NHS Reference Costs for 2010-2011 based on the
information provided by NHS trusts and primary care trusts. The unit cost for acute
day care was not available. However, Curtis (2012) provides unit costs for the day
care in mental health services for different caseload sizes and grades of staff. Acute
day care unit cost was conservatively approximated using day care unit cost
estimate in mental health services assuming that it will be provided by qualified staff
in Band 6 with a caseload of only 10 people resulting in a unit cost of £171. Based on

these crude estimates acute day care could potentially lead to a cost saving of £159
per day of acute care.

The GDG agreed that the main aim of the review of alternatives to acute admission
was to evaluate the feasibility and safety of managing a crisis outside inpatient care,
taking into account service user preference and choice. The GDG also considered
engagement and satisfaction with services to be critical when evaluating this
evidence. Thus, the outcomes considered to be of critical importance were:
• Service use (for example, admission, re-admission)

• Satisfaction with services (service user and carer).
The GDG recognised that no studies adequately dealt with preference and choice.
The GDG took the view that service users should have a range of alternatives to
inpatient care as inpatient care is strongly associated with stigma and considerable
anxiety for service users and their carers.

CRHTTs are a team-based approach to providing treatment and care for people in a
crisis as an alternative in inpatient treatment. The evidence suggests that CRHTTs
reduce admission when compared with standard inpatient care up to 1 year’s
follow-up and possibly up to 2 years’ follow-up. However, there is no evidence of
additional benefit in readmission rates. CRHTTs are probably preferred to inpatient
treatment by service users and they may be superior to inpatient treatment at
engaging service users, as well as improving service user quality of life and clinical
outcomes. In terms of service user choice, the GDG regarded CRHTTs as having
sufficient evidence as an alternative to recommend that these should be available
and should continue to act as the single point of referral for all acute care,
gatekeeping admission to inpatient units.
Acute day hospitals and crisis houses

Acute day hospitals are an alternative to home treatment for a specific service user
group who have support at home in the evening and at night but not during the day;
or as a form of respite for carers. The evidence reviewed here suggests that acute day
hospitals are a viable and clinically effective alternative to inpatient care; and there is
no reason to think that acute day hospitals could not provide evidence based
therapeutic interventions recommended in this guideline. The GDG considered the

acute day hospital to be an important selective alternative to inpatient care generally
preferred by service users.
Crisis houses are an alternative to inpatient admission for service users who do not
have any support at home during the day or in the evenings and night time, or
where carers are unable to cope and/or need respite. The evidence currently
suggests that they may be equivalent to inpatient care, but the evidence reviewed
here is inconclusive. There are a growing number of crisis houses around the UK.
The GDG considered these as a possible alternative to inpatient care if preferred by
service users and an important choice for service users to be able to avoid admission.

Economic evidence on CRHTTs in the UK is based on two studies. Both concluded
that CRHTTs are highly likely to be cost effective when compared with standard
care for people with schizophrenia and other serious mental health problems in an
acute crisis. The cost savings mainly result from the reduction in costs associated
with hospital admissions. The existing economic evidence supports the GDG’s view
that CRHTTs should be offered to all service users as an alternative to inpatient
admission. Although the cost-effectiveness evidence for other alternatives is lacking,
the substantial costs of inpatient treatment make it highly likely that alternatives,
associated with similar or lower costs, would be cost effective.
Acute day hospitals

No economic studies were identified that assessed the cost effectiveness of acute day
hospitals. Nevertheless they were found to be a viable and clinically effective
alternative to inpatient care and an alternative generally preferred by service users.
Moreover, very crude costing indicated that acute inpatient care is associated with
substantial costs and it is highly likely that acute day care would be associated with
similar or lower costs, and would be a cost-effective treatment choice for people with

The quality of the evidence ranged from very low to low across outcomes. Reasons
for downgrading included risk of bias in the included studies, high heterogeneity,
and imprecise confidence intervals. The evidence included in the review of CRHTTs
was of particular concern because of the age of the included trials. This resulted in
possible poor reporting and thus high risk of bias in the included trials.
Additionally, there was serious heterogeneity across the included studies, which
could be explained by the differences in findings between trials from different
countries as UK-only sub-analysis produced more consistent results.
Acute day hospitals and crisis houses

The quality of the evidence base for these reviews ranged from low to high. Reasons
for downgrading included risk of bias, high heterogeneity or lack of precision in
confidence intervals. Heterogeneity was a major concern when evaluating the
evidence. However, although variance was observed in the effect size across studies,
the direction of effect was consistent across most studies. The evidence for crisis
houses was low quality, which was likely to be a result of the lack of available
evidence. The review of acute day hospitals was more robust due to the inclusion of
the large and well-designed EU-multicentre trial. In general, the GDG acknowledged
that although RCTs are an important step in evaluating the impact of complex
interventions such as teams and service-level interventions, there are significant
problems associated with using this type of study design in this context.
The GDG discussed the term ‘acute day hospital’, a now outdated term, and felt this
should be changed to ‘acute day care’ to increase service user choice.
The GDG judged that the evidence supports the recommendation that CRHTTs are a
viable alternative to inpatient admission and should be offered as a first option to
service users in a crisis. Furthermore, the GDG discussed and agreed that CRHTTs
should be the single point of referral and triage for people in a crisis and thus
admission to inpatient care, or any other acute care, should follow assessment by the
CRHTTs. The GDG believed that acute day care, and probably crisis houses, may be
considered as alternatives to inpatient care, justified at least in large part on the basis
of service user preference and to expand choice. The GDG agreed that CRHTTs
should be the cornerstone of acute care in the community, with other alternatives to
inpatient care being determined on the basis of personal circumstances, individual
need and preferences. Following extensive discussion of the acute care pathway in
mental health, the GDG concluded that consideration should be given to the
management of acute care as a whole system or pathway, including CRHTTs, acute
day care, inpatient units and probably crisis houses for those who have no support
at home or in the community. Moreover, other local alternatives such as respite for
service users and for carers should be managed within this local acute care pathway.
Health service managers should also give consideration to the management of the
interface between acute care and non-acute care in the community.
The GDG also considered the impact upon service users of an acute episode of
psychosis or schizophrenia. Service users often understand the experience very
differently from health and social care professionals involved in their care.
Currently, service users’ notes are used predominantly as a record of care and
treatment from the professionals’ perspective. The GDG for the 2014 guideline
agreed with the GDGs for the 2002 and 2009 guidelines that omitting service users’
accounts of their experience introduces systematic bias into the case record and
recommended that service users, especially those who are admitted to hospital,
should add their accounts to their own notes.

12.4.6.1 Offer crisis resolution and home treatment teams as a first-line service to
support people with psychosis or schizophrenia during an acute episode in
the community if the severity of the episode, or the level of risk to self or
others, exceeds the capacity of the early intervention in psychosis services or
other community teams to effectively manage it. [new 2014]
12.4.6.2 Crisis resolution and home treatment teams should be the single point of
entry to all other acute services in the community and in hospitals. [new
2014]
12.4.6.3 Consider acute community treatment within crisis resolution and home
treatment teams before admission to an inpatient unit and as a means to
enable timely discharge from inpatient units. Crisis houses or acute day
facilities may be considered in addition to crisis resolution and home
treatment teams depending on the person’s preference and need. [new 2014]
12.4.6.4 If a person with psychosis or schizophrenia needs hospital care, think about
the impact on the person, their carers and other family members, especially
if the inpatient unit is a long way from where they live. If hospital admission
is unavoidable, ensure that the setting is suitable for the person’s age,
gender and level of vulnerability, support their carers and follow the
recommendations in Service user experience in adult mental health (NICE
12.4.6.5 After each acute episode, encourage people with psychosis or schizophrenia
to write an account of their illness in their notes. [2009]

13 VOCATIONAL REHABILITATION
13.1 INTRODUCTION
This chapter reviews the evidence for vocational rehabilitation interventions and
updates the 2009 guideline. It also includes a new review assessing the efficacy of
cognitive remediation in combination with vocational rehabilitation.
Types of employment vary widely and can mean different things to different people,
for example, it could mean being self-employed, having paid or unpaid employment
(including voluntary work), working part time or in a sheltered environment, or
being in supported employment. A recent estimate of employment for people with
psychosis and schizophrenia is 5 to 15%, with an average of 8% (Schizophrenia
Commission, 2012), which is significantly less than the 71% of the general population
currently employed. Despite much evidence that work has many benefits for people
with psychosis and schizophrenia, the likelihood of employment remains extremely
low. The literature suggests that up to 97.5% of service users may want some type of
work role, for example volunteering or paid employment, but 53% stated they had
not received any support in obtaining work (Seebohm & Secker, 2005).
There are many benefits to having a role in society and performing that role’s
associated tasks (Ross, 2008). Making a contribution to society and promoting
citizenship as a result of a work role can improve recovery (Repper & Perkins, 2003).
It is important to note that without a work role an individual will have limited
income, routines and choices and experience social isolation, which are all
recognised as stressors. Evidence of increased mental distress (reduced self-esteem
and increased psychosomatic symptoms) in the unemployed general population is
widely recorded (Paul & Moser, 2009). The rise in suicide rates with increased
unemployment (Stuckler et al., 2011) reinforces the view that employment can be
better for mental health. Therefore, the right work or vocational role with the right
support can be of great benefit to people with psychosis and schizophrenia in terms
of health, social functioning and financial reward (The Work Foundation, 2013).
However, while recent publications reaffirm the health benefits of open employment
for people with psychosis and schizophrenia (Schizophrenia Commission, 2012; The
Work Foundation, 2013), there is a lack of progress in increasing the numbers in
employment. Many factors contribute to this. Within mental health services, the
negative attitudes of mental health professionals towards people with mental
disorders may lead to pessimism and thus reduce aspirations and the subsequent
provision of services (Hansson et al., 2013). Societal stigma and discrimination, the
diagnostic label, fear of loss of or changes to benefits, and lack of skills in exploring
and putting in place employment support within mainstream services are other
factors that contribute to the problem (Marwaha & Johnson, 2004; The Work
Foundation, 2013).

Guidance to support people with mental illness at work and to manage long-term
sickness absence can be found in public health guidance published by NICE (NICE,
2009b; 2009c).
It is a reasonable assumption that back to work and in work support should be

regarded as an essential element of interventions for people with psychosis and
schizophrenia in recovery (The Work Foundation, 2013), not least because the longer
the period of non-engagement with a role the greater the limitations of such roles
later in life (Bell & Blanchflower, 2011).
Assessment and interventions relating to vocational rehabilitation may be offered by

occupational therapists and specialist employment advisors. To aid speed of access
and a link to other clinical interventions, the person providing employment
interventions is based in the clinical multidisciplinary team.
The predictors for gaining employment for people with psychosis and schizophrenia
are a work history and the desire to work, and there is evidence that the presence of
positive symptoms has a more advantageous influence on work outcomes compared
with negative symptoms (Marwaha & Johnson, 2004). Upon gaining employment, it
is important that people are supported to manage disclosure at work, and negotiate
reasonable adjustments and funding in order to provide the appropriate support to
the employer and employee.
13.2 CLINICAL EVIDENCE REVIEW – VOCATIONAL

REHABILITATION INTERVENTIONS
13.2.1 Introduction
The vocational rehabilitation interventions reviewed in this chapter include standard
and modified supported employment and prevocational training. In addition,
cognitive remediation as a possible adjunct to these interventions is also reviewed.
Cognitive impairment is present in a proportion of people with psychosis
schizophrenia, particularly in the domains of memory (Brenner, 1986), attention
(Oltmanns & Neale, 1975) and executive functions, such as organisation and
planning (Weinberger et al., 1988), and is associated with reduced capacity to work
(Wexler & Bell, 2005). Therefore it is plausible that an intervention designed to
improve cognitive functioning, such as cognitive remediation (Wykes & Reeder,
2005), might also improve performance in employment in people with psychosis and
schizophrenia. It is also possible that vocational rehabilitation programmes might
help people to embed and generalise gains made through previous cognitive
remediation (Wexler & Bell, 2005). The general effectiveness of cognitive remediation
is reviewed in Chapter 9. The current chapter will include a review of the
effectiveness of cognitive remediation when used as an adjunctive treatment to
improve the effectiveness of vocational rehabilitation.

Definitions and aim of interventions
Prevocational training is defined as any approach to vocational rehabilitation in which
participants are expected to undergo a period of preparation before being
encouraged to seek competitive employment. This preparation phase could involve
either work in a sheltered environment (such as a workshop or work unit), or some
form of pre-employment training or transitional employment. This included both
traditional (sheltered workshop) and ‘clubhouse’ approaches.
Supported employment, referred to as individual placement and support (IPS) is any

approach to vocational rehabilitation that attempts to place service users in
competitive employment immediately. It was acceptable for supported employment
to begin with a short period of preparation, but this had to be of less than 1 month’s
duration and not involve work placement in a sheltered setting, training or
transitional employment.
Modifications of vocational rehabilitation programmes are defined as either prevocational

training or supported employment that has been enhanced by some technique to
increase participants’ motivation. Typical techniques consist of payment for
participation in the programme or some form of psychological intervention.
Control is defined as the usual psychiatric care for participants in the trial without
any specific vocational component. In all trials where an intervention was compared
with standard care, unless otherwise stated, participants would have received the
intervention in addition to standard care. Thus, for example, in a trial comparing
prevocational training and standard community care, participants in the former
group would also have been in receipt of standard community services, such as
outpatient appointments.
Cognitive remediation is defined as:

• an identified procedure that is specifically focused on basic cognitive
processes, such as attention, working memory or executive functioning, and
• having the specific intention of bringing about an improvement in the level of
performance on that specified cognitive function or other functions, including
daily living, social or vocational skills.
13.2.2 Clinical review protocol - vocational rehabilitation

interventions
guideline, can be found in Table 151 (the full review protocols and a complete list of

Table 151: Clinical review protocol for the review of vocational rehabilitation
interventions
potential harms of vocational rehabilitation interventions compared with
treatment as usual or another interventions?
Sub-questions a.Supported employment
b. Prevocational training (including individual placement support,
volunteering, training)
c. Modifications of above (paid work or additional psychological therapy)
d. Cognitive remediation with vocational rehabilitation
Objectives To evaluate the effectiveness of vocational rehabilitation interventions for
people with psychosis and schizophrenia.
Population Included
Intervention(s) • Supported employment
• Prevocational training (including individual placement support,
volunteering, training)
• Modifications of above (paid work or additional psychological
therapy)
• Cognitive remediation with vocational rehabilitation
Comparison • Vocational rehabilitation versus any alternative management strategy
• Cognitive remediation and vocational rehabilitration versus
vocational rehabilitation alone
Critical outcomes • Employment and education
o Competitive employment
o Occupation (any non-competitive – for example, volunteer or
unpaid work)
o Attendance at school/college
• Quality of life
Process
Date searched Sub questions a, b, c:
SR/RCT: 2002 to June 2013
Sub question d:
SR: 1995 to June 2013
RCT: database inception to June 2013
NB: Vocational rehabilitation with cognitive rehabilitation was not reviewed

in the 2009 guideline. Therefore, an additional search for SRs/RCTs was run
from an earlier date.

Where more than one follow-up point within the same period were available,
the latest one was reported.
Sub-analysis

studies.
13.2.3 Studies considered 56

The 2009 guideline reviewed vocational rehabilitation interventions alone (without
cognitive remediation), utilising and updating an existing Cochrane review
(Crowther et al., 2001) of 18 RCTs. The Cochrane review was assessed as being up-
to-date by the authors in December 2010. Since then, a number of new trials have
been published and therefore for the 2014 guideline, a new review was conducted.
For the purposes of the guideline, vocational rehabilitation interventions were

categorised as:
• standard supported employment
• modified supported employment (with additional payment or psychological
intervention)
• standard prevocational training
• modified prevocational training (with additional payment or psychological
intervention).
On the basis of the available evidence the reviews conducted involved the following
comparisons:
• supported employment (standard or modified) versus prevocational training

(standard or modified)
• supported employment (standard or modified) versus control (non-
vocational)
• prevocational training (standard or modified) versus control (non-vocational)
• standard prevocational training versus modified prevocational training
• modified prevocational training (paid and psychological intervention) versus
modified prevocational training (paid) supported employment (standard or
modified) plus prevocational training (standard or modified) versus
supported employment alone
• supported employment (standard or modified) plus prevocational training
(standard or modified) versus prevocational training alone

• cognitive remediation with vocational rehabilitation versus vocational
rehabilitation alone.
Vocational rehabilitation alone

Thirty-eight RCTs (N = 8,832) met the eligibility criteria for this review of vocational
rehabilitation interventions: BEARD1963 (Beard et al., 1963), BECKER1967 (Becker,
1967), BELL1993 (Bell et al., 1993), BELL2003 (Bell et al., 2003), BIO2011 (Bio &
Gattaz, 2011) BLANKERTZ1996 (Blankertz & Robinson, 1996), BOND1986 (Bond &
Dincin, 1986), BOND1995 (Bond et al., 1995), BOND2007 (Bond et al., 2007),
BURNS2007 (Burns et al., 2007a), CHANDLER1996 (Chandler et al., 1996),
COOK2005 (Cook et al., 2005), DINCIN1982 (Dincin & Witheridge, 1982),
DRAKE1994 (Drake et al., 1994), DRAKE1999 (Drake et al., 1999), FREY2011 (Frey et
al., 2011), GERVEY1994 (Gervey & Bedell, 1994), GOLD2006 (Gold et al., 2006),
GRIFFITHS1974 (Griffiths, 1974), HOFFMAN2012 (Hoffmann et al., 2012),
HOWARD2010 (Howard et al., 2010), KILLACKEY2008 (Killackey et al., 2008),
KLINE1981 (Kline & Hoisington, 1981), KOPELOWICZ2006 (Kopelowicz et al.,
2006), KULDAU1977 (Kuldau & Dirks, 1977), LATIMER2006 (Latimer et al., 2006),
LEHMAN2002 (Lehman et al., 2002), LYSAKER2005 (Lysaker et al., 2005),
LYSAKER2009 (Lysaker et al., 2009), MCFARLANE2000 (McFarlane et al., 2000),
MUESER2002 57(Mueser et al., 2002a), MUESER2005 (Mueser et al., 2005),
OKPAKU1997 (Okpaku & Anderson, 1997), TSANG2009 (Tsang et al., 2009),
TWAMLEY2012 (Twamley et al., 2012), WALKER1969 (Walker et al., 1969),
WOLKON1971 (Wolkon et al., 1971), WONG2008 (Wong et al., 2008). All 38 studies
were published in peer-reviewed journals between 1963 and 2012. Further
15a. See Table 152, Table 153 and Table 154 for an overview of the trials included in
each category.
Of the eligible trials, 18 included a large proportion (>75%) of participants with a

primary diagnosis of psychosis or schizophrenia. Four of the included trials were
based in the UK/Europe.
57 In the previous guideline MUESER2002 (Mueser et al., 2002) was the conference paper referenced. Since then,
the study data has been published in MUESER2004 (Mueser KT, Clark RE, Haines M, Drake RE, McHugo GJ,
Bond GR, et al. The Hartford study of supported employment for persons with severe mental illness. Journal of
Consulting and Clinical Psychology. 2004;72:479-90.). For the purpose of this guideline and to avoid confusion
the previous study ID of MUESER2002 will be used in this guideline.

Table 152: Study information table for trials comparing vocational rehabilitation interventions with any alternative
management strategy
Supported employment Prevocational training versus TAU Supported employment versus prevocational
versus TAU training
Total no. of trials (k); k = 4; N = 2,687 k = 11; N = 1,598 k = 19; N = 4,192
participants (N)
Study ID CHANDLER1996 BEARD1963 BOND1986
FREY2011 BECKER1967 BOND1995
KILLACKEY2008 BIO2011 BOND2007
OKPAKU1997 BLANKERTZ1996 BURNS2007
DINCIN1982 COOK2005
GRIFFITHS1974 DRAKE1994
KLINE1981 DRAKE1999
KOPELOWICZ2006 GERVEY1994
KULDAU1977 GOLD2006
WALKER1969 HOFFMAN2012
WOLKON1971 HOWARD2010
LATIMER2006
LEHMAN2002
MCFARLANE2000
MUESER2002
MUESER2005
TSANG2009
TWAMLEY2012
WONG2008
Country Australia (k = 1) Brazil (k = 1) Canada (k = 1)
USA (k = 3) UK (k = 1) China (k = 2)
USA (k = 9) Europe (k = 1)
Switzerland (k = 1)
UK (k = 1)
USA (k = 13)
Year of publication 1996 to 2011 1963 to 2011 1986 to 2012
Mean age of 35.19 years (21.36 to 47.4 34.85 years (25.4 to 46 years)2 36.39 years (19 to 51 years)5
participants (range) years)1

Mean percentage of 51.99% (23 to 100%) 75.03% (27.47 to 100%)3 67.71% (38 to 100%)6
participants with
primary diagnosis of
psychosis or
Mean percentage of 39.02% (19.5 to 52.7%) 31.32% (0 to 65%)4 42.25% (20 to 63.79%)
women (range)
Length of treatment 26 to 156 weeks 2 to 78 weeks 8 to 104 weeks
Length of follow-up End of treatment only End of treatment only End of treatment only
CHANDLER1996 BECKER1967 BOND1986
FREY2011 BIO2011 BOND1995
KILLACKEY2008 BLANKERTZ1996 BOND2007
DINCIN1982 BURNS2007
>12 months KULDAU1977 COOK2005
OKPAKU1997 7 WALKER1969 DRAKE1999
GERVEY1994
Up to 6 months GOLD2006
BEARD1963 HOFFMAN2012
KLINE1981 LATIMER2006
KOPELOWICZ2006 LEHMAN2002
MCFARLANE2000
6- 12 months MUESER2002
BEARD1963 TSANG2009
TWAMLEY2012
>12 months WONG2008
BEARD1963
GRIFFITHS1974 6- 12 months
WOLKON1971 HOWARD2010
>12 months
DRAKE1994
MUESER2005
Intervention type Employment-oriented case Community-based hospital industrial Accelerated vocational rehabilitation (k = 1)
management (k = 1) rehabilitation placement (k = 1) Accelerated approach to supported employment (k = 1)
Integrated service agency (k Rehabilitation programme (k = 5) IPS (k = 11)
= 1) Rehabilitation unit (k = 1) ‘Supported employment interventions’ (k = 1)

IPS (k = 1) Thresholds' rehabilitation services (k = Supported employment using job coaches (k = 2)
IPS + TAU (k = 1) 1) Supported employment using natural supports in the
Work experience and discussion group workplace (k = 1)
(k = 1) ACT with IPS (k = 1)
Work-focused programme (k = 1) Family-aided ACT (k = 1)
Work tasks (k = 1) Supported employment (k = 1)
Integrated supported employment (IPS + work-related,
social skills training) (k = 1)
Comparisons Case management services Other community service referral (k = 1) Conventional vocational rehabilitation (k = 3)
from a community mental Usual services (k = 6) Diversified placement approach (k = 1)
health centre (k = 1) Continued treatment programme (k = 1) Enhanced vocational rehabilitation (k = 1)
Usual services (k = 3) Usual ‘Horizon House Incorporated’ Gradual approach to supported employment (k = 1)
services (k = 1) Gradual vocational rehabilitation
Control ward programme (k = 1) Group skills training (k = 1)
Occupational therapy group (k = 1) Prevocational training (k = 1)
Psychosocial rehabilitation and day care programmes
including prevocational training (k = 1)
Psychosocial rehabilitation programme (k = 1)
Sheltered-employment training (k=1)
Standard vocational services (k = 4)
Supported employment + ‘Workplace Fundamentals’
programme (k = 1)
Supported employment program (k = 1)
Traditional vocational rehabilitation programmes (k = 2)
Note. TAU = treatment as usual; IPS = individual placement and support; ACT = assertive community treatment.
1 CHANDLER1996 did not provide data.
2 BEARD1963, GRIFFITHS1974 and WALKER1969 did not provide data.
3 GRIFFITHS1974 did not provide data.
4 BECKER1967, GRIFFITHS1974 and KLINE1981 did not provide data.
5 GOLD2006 did not provide data.
6 GERVEY1994 did not provide data.
7 OKPAKU1997 study had a variable follow-up period. All participants received 4 months of intervention and one 3-month follow-up interview; some were
followed up for as long as 24 months.

management strategy
Modified prevocational training versus standard Modified prevocational training (paid +

prevocational training psychological intervention) versus modified
prevocational training (paid)
Total no. of trials (k); participants (N) k = 2 (N = 354) k = 3 (N = 213)
Study ID BELL1993 BELL2003
MUESER2002 LYSAKER2005
LYSAKER2009
Country USA (k = 2) USA (k = 3)
Mean age of participants (range) 42.24 years (41.23 to 43.25 years) 46.2 years (43.98 to 48.1 years)
Mean percentage of participants with 87.26% (74.51 to 100%) 100% (100 to 100%)
Mean gender (% women) 20.92% (3.62 to 38.21%) 5% (0 to 15%)
Length of treatment 26 to 104 weeks 26 weeks
Length of follow-up End of treatment only End of treatment only
BELL1993 BELL2003
MUESER2002 LYSAKER2005
LYSAKER2009
Intervention type Prevocational training - pay condition (k = 1) Paid work programme + behavioural intervention (k = 1)
Standard vocational services for clients with severe mental Standard support (job placement) + ‘Indianapolis
illness (k = 1) Vocational Intervention Program’ (k = 2)
Comparisons Prevocational training - no pay condition (k = 1) Paid work programme alone (k = 1)

Psychosocial rehabilitation programme (k = 1) Standard support (job placement) (k = 2)

management strategy
Supported employment + prevocational training Supported employment + prevocational training

versus supported employment versus prevocational training
Total no. of trials (k); participants (N) k = 1; N = 163 k = 1; N = 163
Study ID TSANG2009 TSANG2009
Country China (k = 1) China (k = 1)
Year of publication 2009 2009
Mean age of participants (range) 34.56 years 34.56 years
Mean percentage of participants with 75.46% 75.46%
Mean gender (% women) 50.31% 50.31%
Length of treatment 65 weeks 65 weeks
Length of follow-up End of treatment only End of treatment only
TSANG2009 TSANG2009
Intervention type Integrated supported employment (IPS + work-related, Integrated supported employment (IPS + work-related,
social skills training) (k = 1) social skills training) (k = 1)
Comparisons IPS (k = 1) Traditional vocational rehabilitation (k = 1)
Note. IPS = individual placement and support.

Cognitive remediation with vocational rehabilitation
Six RCTs (N = 533) met the eligibility criteria for the review of cognitive remediation
with vocational rehabilitation: BELL2005 (Bell et al., 2005), BELL2008 (Bell et al.,
2008), LINDENMAYER2008 (Lindenmayer et al., 2008), MCGURK2005 (McGurk et
al., 2005), MCGURK2009 (McGurk et al., 2009) VAUTH2005 (Vauth et al., 2005). All
six studies were published in peer-reviewed journals between 2005 and 2009. In
addition, five studies were excluded from the analysis. Further information about
both included and excluded studies can be found in Appendix 15a.
Of the eligible trials, five included a large proportion (>75%) of participants with a
primary diagnosis of psychosis or schizophrenia. None of the included trials were
based in the UK/Europe. Table 155 provides an overview of the trials included in
this review.
Table 155: Study information table for trials comparing cognitive remediation and
vocational rehabilitation interventions with vocational rehabilitation alone
Cognitive remediation with vocational rehabilitation versus

vocational rehabilitation alone
Study ID BELL2005
BELL2008
LINDENMAYER2008
MCGURK2005
MCGURK2009
VAUTH2005
Country Germany (k = 1)
USA (k = 5)
Mean age of participants (range) 39.07 years (28.8 to 44.06 years)
Mean percentage of participants with 87.09% (61.76 to 100%)
Length of treatment 12 to 104 weeks
Length of follow-up End of treatment only
BELL2008
MCGURK2009
Up to 6 months
BELL2005
6- 12 months
LINDENMAYER2008
VAUTH2005
>12 months
MCGURK2005
Intervention type Cognitive remediation programme plus vocational services
programme (k = 1)
Cognitive training (‘Thinking Skills for Work’ programme) plus

supported employment (k = 1)
Computer-assisted cognitive strategy training (plus vocational
rehabilitation (k = 1)
Neurocognitive enhancement therapy plus vocational rehabilitation
(k = 2)
Work programme with cognitive remediation programme (k = 1)
Comparisons Supported employment alone (k = 1)
Vocational rehabilitation alone (k = 2)
Vocational services programme alone (k = 1)
Work programme with computerised control condition (k = 1)
Work therapy alone (k = 1)
13.2.4 Clinical evidence for vocational rehabilitation interventions

Supported employment (standard or modified) versus prevocational
training (standard or modified)
High to moderate quality evidence from up to 18 studies with 3,476 participants
showed that supported employment was more effective than prevocational training
for the outcomes of gaining competitive employment, hours/weeks worked, length
of time in longest job, time to first competitive job, and length of time worked. There
was less conclusive evidence for any benefits regarding duration of employment and
number of jobs held. However, these benefits were found at the end of the
intervention and the longer-term benefits of supported employment over
prevocational training are unclear.
Low to very low quality evidence from up to six studies with 985 participants
suggests that supported employment is more effective than prevocational training in
increasing the chances of placement in any occupation (paid/unpaid/competitive/
uncompetitive), time to obtain any occupation, number of weeks worked and
earnings at the end of the intervention. However, the evidence for effects on the
chances of obtaining a placement in volunteer employment, the number of hours
worked and longest time in one job is inconclusive. None of the included trials
reported follow-up term data and thus the long-term benefits are unclear.
Moderate quality evidence from up to four trials with 699 participants was
inconclusive regarding any benefits on functional disability of either intervention at
the end of the intervention and at medium-term follow-up.
High quality evidence from four studies with 683 participants did not show any
benefit of one intervention over the other in improving quality of life at the end of
the intervention. Longer-term evidence was unavailable.

Table 156: Summary of findings table for trials of supported employment
(standard or modified) compared with prevocational training (standard or
modified)
Patient or population: Adults with psychosis or schizophrenia

Intervention: Supported employment (standard or modified)
Comparison: Pre-vocational training (standard or modified)
Pre- Supported employment (GRADE)
vocational (standard or modified)
training
(standard
or
modified)
Employment Study population RR 0.63 3,627 ⊕⊕⊕⊝
(competitive) - 798 per 1000 503 per 1000 (0.56 to (18 studies) Moderate1
NOT in (447 to 575) 0.72)
competitive
employment, end
of treatment
Employment N/A Mean employment, competitive N/A 2,475 ⊕⊝⊝⊝
(competitive) – – earnings, end of treatment) in (12 studies) Very low2,3
earnings, end of the intervention groups was 0.74
treatment standard deviations higher (0.38
to 1.10 higher)
Employment N/A Mean employment (competitive- N/A 406 ⊕⊕⊝⊝
(competitive) – duration, end of treatment) in (2 studies) Low1,2
duration, end of the intervention groups was 0.17
Employment N/A Mean employment (competitive N/A 661 ⊕⊕⊝⊝
(competitive) - - longest job worked, end of (5 studies) Low1,4
longest job worked, treatment) in the intervention
higher)
Employment N/A Mean employment (competitive N/A 727 ⊕⊕⊕⊕
(competitive) - - time to first job, end of (7 studies) High
time to first job, treatment) in the intervention
lower)
Employment N/A Mean employment (competitive- N/A 221 ⊕⊕⊕⊝
(competitive) - number of jobs, end of (2 studies) Moderate1
number of jobs,- treatment) in the intervention
higher)
Employment N/A Mean employment (competitive N/A 2,404 ⊕⊝⊝⊝
(competitive) - – hours worked, end of (9 studies) Very low2,3
hours worked, end treatment) in the intervention
of treatment

higher)
Employment N/A Mean employment (competitive N/A 994 ⊕⊕⊝⊝
(competitive) - - days/weeks worked, end of (7 studies) Low1,2
days/weeks worked, treatment) in the intervention
higher)
Employment Study population RR 0.92 219 ⊕⊕⊝⊝
(competitive) - 900 per 1000 828 per 1000 (0.82 to (1 study) Low4,5
NOT in (738 to 918) 1.02)
competitive
employment, up to
12 months’ follow-
up
Employment N/A Mean employment (competitive N/A 175 ⊕⊕⊕⊝
(competitive) - – hours worked, >12 months’ (2 studies) Moderate6
hours worked, >12 follow-up) in the intervention
months’ follow-up groups was 0.42 standard
0.91 higher)
Employment N/A Mean employment (competitive N/A 175 ⊕⊝⊝⊝
(competitive) – – earning, >12 months’ follow- (2 studies) Very low2,3,4
earning, >12 up) in the intervention groups
months’ follow-up was 0.37 standard deviations
higher (0.09 lower to 0.84 higher)
Employment N/A Mean employment (competitive N/A 35 ⊕⊕⊕⊝
(competitive) - – number of jobs, >12 months’ (1 study) Moderate4
number of jobs, follow-up) in the intervention
>12 months’ groups was 0.07 standard
follow-up deviations higher (0.59 lower to
0.73 higher)
Employment N/A Mean employment (competitive) N/A 35 ⊕⊕⊕⊝
(competitive) - - days/weeks worked, >12 (1 study) Moderate4
days/ weeks worked months’ follow-up) in the
>12 months’ intervention groups was 0.22
follow-up standard deviations higher (0.44
Occupation (any) - Study population RR 0.70 1,043 ⊕⊝⊝⊝
NOT in any 530 per 1000 371 per 1000 (0.56 to (7 studies) Very low1,2,4
occupation (297 to 461) 0.87)
(paid/unpaid/
531 per 1000 372 per 1000
competitive/
uncompetitive), (297 to 462)
end of treatment
Occupation (any) - Study population RR 1.04 256 ⊕⊕⊝⊝
NOT in volunteer 929 per 1000 966 per 1000 (0.84 to (2 studies) Low1,2
employment, end (780 to 1000) 1.28)
of treatment
870 per 1000 905 per 1000
(731 to 1000)
Occupation (any) - N/A The mean occupation (any - time N/A 494 ⊕⊝⊝⊝
time to first job), to first job, end of treatment) in (4 studies) Very low1,2,4
end of treatment the intervention groups was

(0.42 to 0.05 lower)
Occupation (any) - N/A Mean occupation (any - weeks N/A 731 ⊕⊝⊝⊝
weeks worked, end worked, end of treatment) in the (5 studies) Very low1,2,4
of treatment intervention groups was 0.32
to 0.46 higher)
Occupation (any) - N/A Mean occupation (any - hours N/A 683 ⊕⊕⊝⊝
hours worked, end worked, end of treatment) in the (4 studies) Low1,2
to 0.40 higher)
Occupation (any) - N/A Mean occupation (any - longest N/A 638 ⊕⊕⊝⊝
longest job worked, job worked, end of treatment) in (4 studies) Low1,2
end of treatment the intervention groups was 0.23
standard deviations higher
(0.08 to 0.39 higher)
Occupation (any) - N/A Mean occupation (any - number N/A 186 ⊕⊕⊕⊕
number of jobs, end of jobs, end of treatment) in the (1 study) High
Occupation (any) – N/A Mean occupation (any – N/A 552 ⊕⊕⊝⊝
earnings, end of earnings, end of treatment) in (4 studies) Low1,4
treatment the intervention groups was 0.37
to 0.54 higher)
Global state N/A Mean global state (functional N/A 699 ⊕⊕⊕⊝
(functional disability - end of treatment) in (4 studies) Moderate2
disability) - end of the intervention groups was 0.02
Global state N/A Mean global state (functional N/A 188 ⊕⊕⊕⊝
(functional disability - up to 12 months’ (1 study) Moderate2
disability) - up to follow-up) in the intervention
12 months’ follow- groups was 0.04 standard
up deviations higher (0.25 lower to
0.33 higher)
Quality of life - N/A Mean quality of life (end of N/A 683 ⊕⊕⊕⊕
end of treatment treatment) in the intervention (4 studies) High
0.15 higher)
5 Lack of follow-up data suggests likely publication bias.

Sub-analysis: psychosis and schizophrenia only
For the critical outcomes of competitive employment, the sub-analysis findings did
not differ from the main analysis. Unlike the main analysis, although supported
employment was still superior to prevocational training for the number of people
who obtained any occupation, there was no longer any evidence of a difference
between groups for other proxy measures such as hours worked, earnings, longest
jobs worked, and time to first job. The sub-analysis also did not show any benefit of
either intervention in improving quality of life. No other critical outcome data were
available. See Appendix 16 for the related forest plots.
Sub-analysis: UK/Europe trials only

Unlike the main analysis, there was no evidence in studies based in either the UK or
Europe of a difference between treatment groups in obtaining competitive
employment or in earnings at the end of the intervention. It must be noted that there
was a marked reduction in the number of studies included in this sub-analysis. The
sub-analysis did not differ from the main analysis for the outcomes of hours/weeks
worked and quality of life. No other critical outcome data were available. See
Appendix 16 for the related forest plots.
Supported employment (standard or modified) versus control (non-

vocational)
Three studies with 2,277 participants presented very low quality evidence that
supported employment increased the chance of obtaining competitive employment
at the end of the intervention compared with non-vocational control. However, this
effect was not found at long-term follow-up. One study with 41 participants
provided moderate quality evidence that supported employment increased the
hours worked, however, there was no evidence of a positive effect on
days/weeks/months worked, earnings or time to first job. High quality evidence
from one study with 2,055 participants showed that supported employment was
superior to non-vocational control on quality of life and occupational employment
outcomes such as obtaining occupation, days/weeks/months worked, earnings,
hours worked per week, and highest hourly wage. No functional disability data
were available. See Appendix 16 for the related forest plots.

For the critical outcomes related to competitive employment, the sub-analysis
findings did not differ from the main analysis. No other critical outcome data were
available. See Appendix 16 for the related forest plots.

Table 157: Summary of findings table for trials of supported employment
(standard or modified) compared with control (non-vocational)

Intervention: Supported employment (standard or modified)
Comparison: TAU/Control (non-vocational comparison group)
TAU/control Supported employment (GRADE)
(non- (standard or modified)
vocational
comparison
group)
Employment Study population RR 0.46 2,277 ⊕⊝⊝⊝
(competitive) - NOT in 687 per 1000 316 per 1000 (0.25 to (3 studies) Very low1,2,3
competitive (172 to 584) 0.85)
employment, end of
849 per 1000 391 per 1000
treatment
(212 to 722)
Employment N/A Mean employment N/A 41 ⊕⊕⊕⊝
(competitive) - days/ (competitive - days/ weeks/ (1 study) Moderate3
weeks/ months worked, months worked, end of
end of treatment treatment) in the intervention
deviations higher (1.11 lower
to 0.13 higher)
(competitive) - hours (competitive - hours worked, (1 study) Moderate4
worked, end of end of treatment) in the
treatment intervention groups was 0.85
(0.20 to 1.49 higher
(competitive) – (competitive – earnings, end (1 study) Moderate3
earnings, end of of treatment) in the
Employment N/A Mean employment N/A 873 ⊕⊕⊕⊕
(competitive) - time to (competitive - time to first job (1 study) High
first job, end of - end of treatment) in the
standard deviations lower
Employment Study population RR 0.76 152 ⊕⊝⊝⊝
(competitive) - NOT in 646 per 1000 491 per 1000 (0.57 to (1 study) Very low3,5,6
employment, > 12
646 per 1000 491 per 1000
months’ follow-up
(368 to 659)
Occupation (any) - Study population RR 0.67 2,055 ⊕⊕⊕⊕
NOT in any 598 per 1000 400 per 1000 (0.61 to (1 study) High
occupation, end of (364 to 436) 0.73)

treatment 598 per 1000 401 per 1000
(365 to 437)
Occupation (any) - N/A Mean occupation (any- time N/A 1,028 ⊕⊕⊕⊕
time to first job, end of to first job, end of treatment) (1 study) High
treatment in the intervention groups
lower (0.24 lower to 0.01
higher)
Occupation (any) - N/A Mean occupation (any - N/A 2,055 ⊕⊕⊕⊕
days/weeks/months days/weeks/months worked, (1 study) High
(0.28 to 0.46 higher)
Occupation (any) - N/A Mean occupation (any- N/A 2,055 ⊕⊕⊕⊕
weekly earnings, end weekly earnings, end of (1 study) High
of treatment treatment) in the intervention
higher)
Occupation (any) - N/A Mean occupation (any - past 3 N/A 2,055 ⊕⊕⊕⊕
past 3 months’ months’ earnings, end of (1 study) High
earnings, end of treatment) in the intervention
higher)
Occupation (any) - N/A Mean occupation (any - hours N/A 2,055 ⊕⊕⊕⊕
hours per week, end of per week, end of treatment) (1 study) High
treatment in the intervention groups
higher (0.28 to 0.45 higher)
Occupation (any) - N/A Mean occupation (any - N/A 2,055 ⊕⊕⊕⊕
highest hourly wage, highest hourly wage, end of (1 study) High
higher)
Quality of life - end of N/A Mean quality of life (end of N/A 2,055 ⊕⊕⊕⊕
treatment treatment) in the intervention (1 study) High
lower)
Note. TAU = treatment as usual; CI = confidence interval; RR = risk ratio.
estimate of effect.
6 Intervention and sample may not be representative.

Prevocational training (standard or modified) versus control (non-
vocational)
There was no evidence that prevocational training was more effective than non-
vocational control in obtaining competitive employment (both at the end of
treatment and at follow-up) or increasing earnings. However, five studies with 641
participants presented very low quality evidence that prevocational training was
effective in obtaining any occupation at the end of treatment. There was however no
evidence for this effect at short- and long-term follow-up. In addition, a very small
study (28 participants) also provided very low quality evidence of an increase in
hours worked for the prevocational intervention compared with non-vocational
control. There was no conclusive evidence of any benefits on attendance in education
at the end of treatment.
Moderate quality evidence from one study (N = 91) shows that prevocational
training is more effective than non-vocational control in increasing quality of life.
This was found at the end of the intervention and follow-up evidence was not
available. No functional disability data were available.

For the critical outcome of competitive employment and quality of life, the sub-
analysis findings did not differ from the main analysis. However, there was no
longer evidence of any benefit of prevocational training for occupation-related
outcomes. No other critical outcome data were available. See Appendix 16 for the
related forest plots.
Sub-analysis: UK/Europe trials only

As with the main analysis, there was no evidence that prevocational training was
more effective than non-vocational control in obtaining competitive employment at
follow-up. No other critical outcome data were available. See Appendix 16 for the
related forest plots.

Table 158: Summary of findings table for prevocational training (standard or
modified) compared with control (non-vocational)

Intervention: Prevocational training (standard or modified)
Comparison: TAU/active control (non-vocational comparison group)
Outcomes Illustrative comparative risks* (95% CI) Relative No of Quality of
Assumed risk Corresponding risk effect Participants the
TAU/Active Prevocational training (95% CI) (studies) evidence
control (non- (standard or modified) (GRADE)
vocational
comparison
group)
(competitive) - 766 per 1000 667 per 1000 (0.76 to (5 studies) Low1,2
NOT in (582 to 774) 1.01)
competitive
688 per 1000 599 per 1000
employment, end
of treatment (523 to 695)
(competitive) – (competitive – earnings, end of (1 study) Moderate3
earnings, end of treatment) in the intervention
deviations higher (0.16 lower
to 0.68 higher)
(competitive) - up 786 per 1000 927 per 1000 (0.87 to (1 study) Low3,4
to 12 months’ (684 to 1000) 1.61)
follow-up
786 per 1000 927 per 1000
(684 to 1000)
Occupation (any) Mean occupation (any - hours 28 ⊕⊕⊝⊝
- hours worked, worked, end of treatment) in (1 study) Low2,3
end of treatment the intervention groups was
(0.03 to 1.58 lower)
Occupation (any) Study population RR 0.73 641 ⊕⊝⊝⊝
- NOT in any 819 per 1000 598 per 1000 (0.58 to (5 studies) Very low1,2,5
occupation, end of (475 to 761) 0.93)
treatment
786 per 1000 574 per 1000
(456 to 731)
Occupation (any) Study population RR 0.78 268 ⊕⊝⊝⊝
- up to 6 months’ 803 per 1000 626 per 1000 (0.53 to (2 studies) Very low1,2,4,5
follow-up (425 to 915) 1.14)
843 per 1000 658 per 1000
(447 to 961)
Occupation (any)- Study population RR 0.88 215 ⊕⊝⊝⊝
NOT employed, 7- 750 per 1000 660 per 1000 (0.72 to (1 study) Very low2,3,4
12 months’ follow- (540 to 795) 1.06)
up
750 per 1000 660 per 1000
(540 to 795)
Education Study population RR 0.94 211 ⊕⊕⊕⊝

(attendance) - 936 per 1000 880 per 1000 (0.88 to (2 studies) Moderate1
NOT attending, (823 to 945) 1.01)
end of treatment 927 per 1000 871 per 1000
(816 to 936)
Quality of life - Mean quality of life (end of 91 ⊕⊕⊕⊝
end of treatment treatment) in the intervention (1 study) Moderate3
lower)
Note. TAU = treatment as usual; CI = confidence interval; RR = risk ratio;
*The basis for the assumed risk (for example, the median control group risk across studies) is
provided in the footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the
3 Crucial limitation for one criterion or some limitations for multiple criteria sufficient to lower
confidence in the estimate of effect.

5 Evidence of serious heterogeneity of study effect size
Modified prevocational training versus standard prevocational training

There was no evidence of any difference between standard and modified
prevocational training in obtaining competitive employment earnings, hours
worked, and duration of longest job worked at the end of treatment. Moderate
quality evidence from one study with 136 participants showed that standard
prevocational training was effective at increasing the number of weeks worked, but
modified prevocational training was more effective for the outcome of time to first
job at the end of the intervention.
Two studies with 286 participants presented very low to moderate quality evidence
that modified prevocational training was more effective than standard prevocational
training for obtaining any occupation, earnings, hours worked and time to first job at
the end of the intervention. Follow-up data were not available. There was no
evidence of any difference between modified and standard prevocational training in
terms of weeks worked and longest job worked in any occupation. No functional
disability or quality of life data were available.

For the critical outcomes associated with competitive employment and occupation,
the sub-analysis findings did not differ from the main analysis. No other critical

Table 159: Summary of findings table for trials of modified prevocational training
compared with standard prevocational training

Intervention: Modified prevocational training
Comparison: Standard prevocational training
Assumed risk Corresponding risk effect participants the
Standard Modified prevocational (95% CI) (studies) evidence
prevocational training (GRADE)
training
(competitive) - NOT in 821 per 1000 722 per 1000 (0.73 to (1 study) Low1,2
employment, end of
544 per 1000 479 per 1000
treatment
(397 to 577)
(competitive)- earnings, (competitive – earnings, end of (1 study) Moderate1
0.58 higher)
(competitive) - weeks (competitive - weeks worked, (1 study) Moderate1
Employment N/A Mean employment N/A 136 ⊕⊕⊝⊝
(competitive) - hours (competitive - hours worked, (1 study) Low1,2
Employment N/A Mean employment N/A 136 ⊕⊕⊝⊝
(competitive) - longest (competitive - longest job (1 study) Low1,2
job worked, end of worked, end of treatment) in
(competitive) - time to (competitive - time to first job, (1 study) Moderate1
first job, end of end of treatment) in the
(1.1 to 0.42 lower)
Occupation (any) - Study population RR 0.53 286 ⊕⊝⊝⊝
NOT in any paid (0.3 to (2 studies) Very
(competitive or 708 per 1000 375 per 1000 0.94) low1,2,3
uncompetitive) (212 to 666)
employment, end of 300 per 1000 159 per 1000
treatment (90 to 282)
Occupation (any) – N/A Mean occupation (any – N/A 280 ⊕⊝⊝⊝

earnings, end of earnings, end of treatment) in (2 studies) Very
treatment the intervention groups was low1,4
(0.46 to 0.95 higher)
Occupation (any) - N/A Mean occupation (any - weeks N/A 136 ⊕⊕⊝⊝
weeks worked, end of worked, end of treatment) in (1 study) Low1,2
Occupation (any) - N/A Mean occupation (any - hours N/A 280 ⊕⊕⊕⊝
hours worked, end of worked, end of treatment) in (2 studies) Moderate1
(0.58 to 1.21 lower)
Occupation (any) - N/A Mean occupation (any - longest N/A 136 ⊕⊕⊝⊝
longest job worked, end job worked, end of treatment) (1 study) Low1,2
of treatment in the intervention groups was
Occupation (any) - time N/A Mean occupation (any - time to N/A 136 ⊕⊕⊝⊝
to first job, end of first job, end of treatment) in (1 study) Low1,2
(0.95 to 0.25 lower)
estimate of effect.
Modified prevocational training (paid and psychological intervention)

versus modified prevocational training (paid)
Low quality evidence from up to three studies with 210 participants showed that
modifying prevocational training with both payment and the addition of a
psychological intervention component was more effective than payment alone for
the number of weeks worked and the number of hours worked in any occupation,
and quality of life at the end of the intervention period. No other employment-
related or quality of life outcomes were available.
presented in Table 160Error! Not a valid bookmark self-reference.. The full
evidence profiles and associated forest plots can be found in Appendix 17 and
Appendix 16, respectively.

The sub-analysis findings did not differ from the main analysis. See Appendix 16 for
the related forest plots.

Table 160: Summary of findings table for modified prevocational training (paid
and psychological intervention) compared with modified prevocational training
(paid)

Intervention: Modified prevocational training (paid + psychological intervention)
Comparison: Modified prevocational training (+ paid)
Assumed risk Corresponding risk effect participants the
Modified Modified prevocational (95% CI) (studies) evidence
prevocational training (paid + psycho- (GRADE)
training (+paid) logical intervention)
Occupation (any) - N/A Mean occupation (any - N/A 147 ⊕⊕⊝⊝
weeks worked, end weeks worked, end of (2 studies) Low1,2
of treatment treatment) in the
intervention groups was
0.51 standard deviations
Occupation (any) - N/A Mean occupation (any - N/A 147 ⊕⊕⊝⊝
hours worked,- end hours worked, end of (2 studies) Low2
of treatment treatment) in the
intervention groups was
Functional N/A Mean functional disability N/A 210 ⊕⊕⊝⊝
disability - end of (end of treatment) in the (3 studies) Low3
1 Most of the information is from studies at moderate risk of bias.
Supported employment plus prevocational training versus supported

employment alone
Moderate quality evidence from one study with 107 participants showed that a
combined supported employment and prevocational training intervention was more
effective than supported employment alone in obtaining competitive employment
and earnings at the end of the intervention. No other critical outcome data were
available.

Table 161: Summary of findings table supported employment plus prevocational
training compared with supported employment alone

Intervention: Supported employment plus prevocational training
Comparison: Supported employment
Outcomes Illustrative comparative risks* Relative No. of Quality of the
(95% CI) effect participants evidence
Assumed Corresponding risk (95% CI) (studies) (GRADE)
risk
Supported Supported
employment employment +
prevocational
training
Employment Study population RR 0.46 108 ⊕⊕⊕⊝
(competitive) - 464 per 1000 214 per 1000 (0.25 to (1 study) Moderate1
end of (116 to 385) 0.83)
treatment
Employment, N/A Mean employment, N/A 108 ⊕⊕⊕⊝
(competitive) – (competitive - (1 study) Moderate2
earnings, end of earnings, end of
treatment treatment) in the
intervention groups
was 0.34 standard
deviations higher
(0.04 lower to 0.72
higher)
*The basis for the assumed risk (for example, the median control group risk across studies) is
provided in the footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the
Supported employment plus prevocational training versus prevocational

training
Moderate quality evidence from one study with 108 participants showed that a
combined supported employment and prevocational training intervention was more
effective than prevocational training alone in obtaining competitive employment at
the end of the intervention. There was no evidence of any difference between groups
in earnings. No other critical outcome data were available.

Table 162: Summary of findings table for supported employment plus
prevocational training compared with prevocational training alone

Intervention: Supported employment + prevocational training
Comparison: Prevocational training
Outcomes Illustrative comparative risks* (95% Relative No. of Quality of
CI) effect participants the evidence
Prevocational Supported employment
training + prevocational training
Employment Study population RR 0.23 107 ⊕⊕⊕⊝
(competitive) - 927 per 1000 213 per 1000 (0.13 to 0.39) (1 study) Moderate1
end of treatment (121 to 362)
Employment, N/A Mean employment, N/A 107 ⊕⊕⊕⊝
(competitive) – (competitive – earnings, (1 study) Moderate1
earnings, end of end of treatment) in the
1 Optimal information size not met
Cognitive remediation with vocational rehabilitation versus vocational

rehabilitation alone
Low quality evidence from two studies with 116 participants showed that combined
vocational rehabilitation and cognitive remediation was more effective than
vocational rehabilitation alone for gaining competitive employment at the end of the
intervention. However, there was no evidence of a benefit at short- and medium-
term follow-up. There was no conclusive evidence of any added benefit on the
outcomes of hours/weeks worked, number of jobs or earnings at the end of the
intervention. No further follow-up data were available. Data assessing rates of
obtaining any occupation at the end of treatment were unavailable.
Very low quality evidence from one study with 34 participants showed that the
combined intervention was more effective than control for the outcome of weeks
worked in any occupation (maintained when assessed at medium-term follow-up).
However, the evidence for any benefit of cognitive remediation with vocational
rehabilitation on hours worked or earnings in any occupation were inconclusive
across follow-up time points. No other critical outcome data were available.

Table 163: Summary of findings table for cognitive remediation with trials of vocational rehabilitation (all) with cognitive
rehabilitation compared with vocational rehabilitation alone

Intervention: Cognitive remediation + vocational rehabilitation
Comparison: Vocational rehabilitation
Outcomes Illustrative comparative risks* (95% CI) Relative No. of participants Quality of the
Assumed risk Corresponding risk effect (studies) evidence
Vocational Cognitive remediation + vocational rehabilitation (95% CI) (GRADE)
rehabilitation
Employment (competitive) - NOT in Study population RR 0.47 116 ⊕⊝⊝⊝
competitive employment, end of 745 per 1000 350 per 1000 (0.24 to 0.92) (2 studies) Very low1,2,3
treatment (179 to 686)
Employment (competitive) - hours N/A Mean employment (competitive - hours worked, end of N/A 150 ⊕⊝⊝⊝
worked, end of treatment treatment) in the intervention groups was 0.38 standard (3 studies) Very low1,3
Employment (competitive) - number N/A Mean employment (competitive- number of jobs, end of N/A 116 ⊕⊝⊝⊝
of jobs, end of treatment treatment) in the intervention groups was 0.57 standard (2 studies) Very low1,2,3
Employment (competitive) - weeks N/A Mean employment (competitive- weeks worked, end of N/A 106 ⊕⊕⊝⊝
worked, end of treatment treatment) in the intervention groups was 0.05 standard (2 studies) Low1,3
deviations lower (0.33 lower to 0.43 higher)
Employment (competitive) – N/A Mean employment (competitive – earnings, end of treatment) N/A 78 ⊕⊝⊝⊝
earnings, end of treatment in the intervention groups was 0.54 standard deviations (2 studies) Very low1,2,3
Employment (competitive) - NOT in Study population RR 0.90 127 ⊕⊕⊝⊝
competitive employment, up to 6 761 per 1000 685 per 1000 (0.72 to 1.12) (1 study) Low4,5
months’ follow-up (548 to 853)
Employment (competitive) - NOT in Study population RR 0.61 65 ⊕⊕⊝⊝
competitive employment, up to 12 571 per 1000 349 per 1000 (0.36 to 1.06) (1 study) Low3,4
months’ follow-up (206 to 606)
Occupation (any) - hours worked, end N/A Mean occupation (any - hours worked, end of treatment) in N/A 233 ⊕⊝⊝⊝

of treatment the intervention groups was 0.02 standard deviations lower (3 studies) Very low1,2,3
Occupation (any) – earnings, end of N/A The mean occupation (any – earnings, end of treatment) in the N/A 161 ⊕⊝⊝⊝
treatment intervention groups was 0.23 standard deviations higher (0.70 (2 studies) Very low1,2,3
Occupation (any) - weeks worked, N/A Mean occupation (any - weeks worked, end of treatment) in N/A 34 ⊕⊕⊝⊝
end of treatment the intervention groups was 0.89 standard deviations higher (1 study) Low3,4
Occupation (any) - hours worked, up N/A Mean occupation (any - hours worked, up to 6 months’ N/A 127 ⊕⊕⊝⊝
to 6 months’ follow-up follow-up) in the intervention groups was 0.45 higher (0.1 to (1 study) Low3,4
0.8 higher)
Occupation (any) - earnings, up to 6 N/A Mean occupation (any – earnings, up to 6 months’ follow-up) N/A 127 ⊕⊕⊝⊝
months’ follow-up in the intervention groups was 0.14 standard deviations (1 study) Low3,4
Occupation (any) - did not obtain Study population RR 0.75 68 ⊕⊕⊕⊝
work, up to 12 months’ follow-up 645 per 1000 484 per 1000 (0.49 to 1.15) (1 study) Moderate3
(316 to 742)
Occupation (any) - hours worked, up N/A Mean occupation (any - hours worked, up to 12 months’ N/A 68 ⊕⊕⊕⊝
to 12 months’ follow-up follow-up) in the intervention groups was 0.43 standard (1 study) Moderate3
Occupation (any) - weeks worked, up N/A Mean occupation (any - weeks worked, up to 12 months’ N/A 68 ⊕⊕⊕⊝
to 12 months’ follow-up follow-up) in the intervention groups was 0.49 standard (1 study) Moderate3
Occupation (any) – earnings, up to N/A Mean occupation (any – earnings, up to 12 months’ follow- N/A 68 ⊕⊕⊕⊝
12 months’ follow-up up) in the intervention groups was 0.39 standard deviations (1 study) Moderate3
*The basis for the assumed risk (for example, the median control group risk across studies) is provided in the footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the
4 Crucial limitation for one criterion or some limitations for multiple criteria sufficient to lower confidence in the estimate. of effect

Overall, the clinical evidence suggests that supported employment is the most
effective vocational rehabilitation method for obtaining competitive employment
and for obtaining any occupation (paid/unpaid or voluntary). Furthermore, there is
consistent evidence across a number of outcome measures that supported
employment is more effective than prevocational training in increasing competitive
employment. Evidence regarding earnings and being able to sustain employment or
any occupation is less conclusive. Additionally, the long-term benefits of supported
employment are not known. This was also found to be the case for sub-analyses
using the studies with a high proportion of participants with psychosis and
schizophrenia. However, this finding was no longer apparent for UK/Europe-based
studies although caution must be exercised when interpreting the results due to the
smaller number of studies eligible for these sub-analyses. Evidence regarding
functional disability and quality of life was less conclusive and no firm conclusions
could be drawn from the available evidence. Findings from a single study showed
that a combination of supported employment with prevocational training was more
effective than either prevocational training or supported employment alone in
gaining competitive employment at the end of treatment but long-term efficacy is
unknown.
Although prevocational training was not found to increase the chances of obtaining
competitive employment, it was beneficial for obtaining any occupation. However,
again, there was no evidence of any benefit beyond the conclusion of the
intervention and this finding was no longer apparent in sub-analyses including only
psychosis and schizophrenia samples. The UK/Europe sub-analysis did not differ
from the main findings. Prevocational training was however found to improve
quality of life but this was on the basis of a single small study.
Modifications to prevocational training via payment or the addition of a

psychological intervention was not additionally beneficial for obtaining competitive
employment. It was however beneficial for obtaining any occupation, speed of
gaining occupation, increasing earnings and job retention although long-term
benefits are not known. The combined modification of a psychological intervention
and payment with prevocational training was found to be more beneficial than
payment alone for the number of hours/weeks worked in any occupation. This was
also the case in the psychosis and schizophrenia diagnosis sub-analysis. However
findings are based on only two studies and the effects in the long-term are unknown.
Lastly, the combined intervention of vocational rehabilitation (any type) with

cognitive remediation was found to be effective for obtaining employment at the end
of the intervention period. However, this outcome was based on a single study and
no further longer-term benefits were found. There was no benefit of the combined
intervention on other proxy vocational outcome measures such as earnings,
hours/weeks worked and number of jobs. In addition, the evidence for obtaining
any occupation was inconclusive showing benefit for the combined intervention at

some follow-up points but not others. The same was found in the psychosis and
schizophrenia sub-analysis.
13.3 HEALTH ECONOMICS EVIDENCE

13.3.1 Systematic literature review
The systematic literature search identified one eligible UK study (Heslin et al., 2011;
Howard et al., 2010), one international study reporting outcomes for the UK (Knapp
et al., 2013) and one US study (Dixon et al., 2002). Details on the methods used for
the systematic search of the economic literature are described in Chapter 3.
References to included studies and evidence tables for all economic studies included
in the guideline systematic literature review are presented in Appendix 19.
Completed methodology checklists of the studies are provided in Appendix 18.
Economic evidence profiles of studies considered during guideline development
(that is, studies that fully or partly met the applicability and quality criteria) are
presented in Appendix 17, accompanying the respective GRADE clinical evidence
profiles.
The UK study was based on an RCT (HOWARD2010) (n = 219) and evaluated the
cost effectiveness of supported employment compared with standard care that
consisted of existing psychosocial rehabilitation, day care programmes and
prevocational training. Howard and colleagues (2010) reported outcomes at 1-year
follow-up and Heslin and colleagues (2011) at 2-year follow-up. The analysis
included intervention costs and the costs of primary, secondary and community
care. The intervention was provided by a not-for-profit, non-governmental
supported employment agency with the support provided by CMHTs. The mean
cost of intervention per person over 2 years was estimated to be approximately £300
in 2006/07 prices. Supported employment resulted in cost savings at 1- and 2-year
follow-up of £2,176 (p < 0.05) and £2,361 (p = ns), respectively. Also, supported
employment resulted in better vocational outcomes at years 1 and 2 (risk ratio of 1.35
[95%CI: 0.95; 1.93] and 1.91 [95%CI: 0.98; 3.74], respectively). However, these
differences were statistically non-significant. Only when authors controlled for all
sociodemographic factors and clinical measures at baseline did results reach
statistical significance at year 1. Nevertheless, the authors concluded that even
though supported employment was a dominant strategy based on point estimates,
the overall benefits were modest and additional interventions may need to be
provided to promote social inclusion for the majority of individuals with severe
mental illness. The above cost-effectiveness analysis was judged to be directly
applicable to this guideline review and the NICE reference case. However, the
analysis was based on a single RCT conducted in south London which may limit the
generalisability of the findings. Also, the components of the intervention and
standard care were not well reported. Moreover, the intervention cost of £339 (in
2011/12 prices) associated with the provision of a supported employment
programme seems to be very low when compared with the unit cost ranging from as
high as £7,188 to £1,902 (depending on the caseload and the provider of the
intervention) as reported by Curtis (2012). According to the authors, the supported

employment intervention was not optimally provided in the RCT and other authors
have expressed concerns about the fidelity of the IPS service delivered (Latimer,
2010). According to Latimer (2010) vocational workers had far fewer contacts with
clients and employers than normal and its hardly surprising that an intervention of
such low intensity had little or no effects. Based on the above considerations the
analysis was judged by the GDG to have potentially serious methodological
limitations.
Knapp and colleagues (Knapp et al., 2013) conducted a cost-effectiveness analysis

comparing IPS with standard care over 18 months. This economic evaluation was
based on an international trial (BURNS2007) (n = 312). The sample was drawn from
six European cities: Groningen (Netherlands), London (UK), Rimini (Italy), Sofia
(Bulgaria), Ulm-Günzburg (Germany) and Zurich (Switzerland). Standard care
varied across sites and consisted of the best typical vocational rehabilitation services
in each city, followed the train-and-place approach and consisted of day treatment in
all cities except for residential care in Ulm-Günzburg. The study population
comprised individuals with severe mental illness including schizophrenia and
schizophrenia-like disorders, bipolar disorder, or depression with psychotic features.
The analysis was conducted from the perspective of health and social care and
included costs associated with intervention provision, accommodation, inpatient and
outpatient services, community-based services, community-based professions and
medication. The outcome measures were the number of days worked in competitive
settings and the percentage of sample members who worked at least 1 day. The
analysis reported pooled results and results for individual sites. In the RCT it was
found that at 18 months 55% of individuals assigned to IPS worked at least 1 day
during the 18-month follow-up period compared with 28% individuals assigned to
vocational services. Moreover, in the UK total 18-month costs per person were £7,414
and £10,985 in the IPS and vocational services groups respectively (in 2003 prices),
resulting in savings of £3,769 (p<0.05). The authors did not report the number of
days worked in competitive settings. Nevertheless, it was found that IPS was
dominant when compared with vocational services using both outcomes in all sites
except at Groningen, where IPS resulted in an additional cost of £30 per person for
an additional 1% of individuals working at least 1 day in a competitive setting and
an additional £10 per person for an additional day of work. Cost-effectiveness
acceptability curves (CEACs) indicated that at a willingness to pay of £0-£1,000 for
an additional 1% of clients working for at least 1 day over the 18-month period, or
for an additional day of work, the probability of IPS being cost effective when
compared with vocational services was nearly equal to 1.00. The authors have
further attempted a partial cost-benefit analysis where intervention costs and the
monetary value of employment were considered. According to the analysis, IPS was
associated with a net benefit of £17,005. The authors concluded that IPS represents a
more efficient use of resources than standard care. Overall this study was judged to
be directly applicable to this guideline review and the NICE reference case, since it
reported a sub-analysis for the UK (London). In the RCT only a small proportion of
the sample was based in the UK (n = 50). Nevertheless, the pattern of the main
findings was consistent across all sites except Groningen, where according to the

authors IPS was implemented in the least effective way. The use of the percentage of
sample members who worked at least one day as an outcome may have potentially
biased results towards IPS. However, IPS was found to be dominant using the
number of days worked in competitive settings as an outcome and also IPS was
associated with the net benefit of £17,005. Although the analysis did not include
QALYs it was not a problem since the intervention was found to be dominant in the
UK. The time frame of the analysis was under 2 years, which may not be sufficiently
long enough to capture the full effects of the intervention. Nevertheless, overall this
was a well-conducted analysis and was judged by the GDG as having only minor
Finally, Dixon and colleagues (2002) assessed the cost effectiveness of supported
employment compared with standard care in service users with schizophrenia,
schizoaffective disorder, bipolar disorder, recurrent major depression or borderline
personality disorder. Standard care was defined as an enhanced vocational
rehabilitation programme. The analysis was based on an RCT (n = 152)
(DRAKE1999) conducted in the US from the public sector perspective. The time
horizon of the analysis was 18 months. The authors found that supported
employment led to a cost increase of $3,968 and resulted in significantly greater
number of hours/weeks of competitive work; however standard care was associated
with greater combined earnings. Consequently, supported employment was
associated with additional costs of $13 and $283 per extra hour and week of
competitive work, respectively, and was dominated by standard care when
combined earnings were used as an outcome. As a result, the authors were unable to
reach any firm conclusions pertaining to the cost effectiveness of supported
employment. The above cost analysis was judged to be only partially applicable to
this guideline review and the NICE reference case. The time horizon of the analysis
was under 2 years, which may not be sufficiently long enough to capture the
outcomes associated with the intervention. Overall the analysis was well conducted
and was judged by the GDG to have only minor methodological limitations.
13.3.2 Economic modelling

Introduction – objective of economic modelling
Provision of supported employment programmes in adults with psychosis and
schizophrenia is an area with potentially major resource implications. The UK study
by Howard and colleagues (2010) had potentially serious methodological limitations
due to sub-optimal provision of IPS and the study by Knapp and colleagues (2013)
was a multi-centre RCT with only 50 participants from the UK site. Consequently, an
economic model was developed to assess the potential cost effectiveness of these
programmes for this population. Supported employment programmes may be
delivered by a range of different providers including health, social care and third
sector organisations. The economic analysis considered IPS and used resource use
estimates from the perspective of the NHS and personal social services (PSS), as
reported in Curtis (2012). UK clinical evidence on supported employment
programmes was very limited, consequently clinical data for the economic analysis

are derived from international RCTs including CHANDLER1996, FREY2011 and
KILLACKEY2008, which compared a supported employment programme with
treatment as usual (TAU) and reported the number of participants who found paid
employment in each group following the supported employment programme.
Economic modelling methods
Interventions assessed
The model was developed to assess the cost effectiveness of a supported
employment programme compared with TAU. The service content of supported
employment and the definition of TAU varied across the studies. In
CHANDLER1996 the supported employment programme was provided by
multidisciplinary teams. The programme was part of integrated services comprising
ACT. TAU was described as local mental health services comprising limited case
management and other rehabilitative services. In FREY2011 the supported
employment programme was part of integrated services that comprised access to
supported employment and systematic medication management services. The
programme focused on consumer choice, integrated services, competitive
employment in regular work settings, rapid job search, personalised follow-on
support, person-centred services and benefits counselling. TAU included a
comprehensive range of services available in the local community that were sought
out by the service user and may have included employment. In KILLACKEY2008 the
supported employment programme was provided in combination with TAU. The
vocational intervention was provided by an employment consultant enlisted for the
project. TAU consisted of care from an Early Psychosis Prevention and Intervention
Centre (EPPIC) that included individual case management, medical review and
referral to external vocational agencies, as well as involvement with the group
programme at EPPIC, which may involve participation in the vocationally-
orientated groups within the programme. TAU was delivered primarily by EPPIC
case managers.
As is clear from the descriptions above, TAU comprised a wide range of

interventions, which were difficult to combine in terms of relevant resource use for
the purposes of economic modelling. Also, the reported information on the resource
utilisation in the studies was not adequate to allow costing. Consequently for the
purposes of the economic model, TAU was defined as day services, which is
reported as an alternative to supported employment in the UK in Curtis (2012).
Model structure
A simple decision-tree followed by a two-state Markov model was constructed using
Microsoft Excel XP in order to assess the costs and outcomes associated with
provision of supported employment and TAU in adults with psychosis and
schizophrenia actively seeking employment. The economic model is an adaptation of
the economic model that assessed supported employment versus standard care (day
services) in people with autism that was developed for the NICE clinical guideline
on autism in adults (NICE, 2012a).

According to the decision-tree model, which was based on the data reported in
CHANDLER1996, FREY2011 and KILLACKEY2008, interventions were provided
over a mean of 22 months. Over this period the mean length of time spent in
employment was estimated to be 10.75 months in the intervention group versus
10.37 months in the TAU groups. Subsequently, a simple Markov model was
developed to estimate the number of adults remaining in employment every year
from endpoint of the decision-tree (that is, from the end of provision of the
intervention) and up to 10 years, using an estimated 10-year job retention rate in
those who found employment following the intervention. The Markov model
consisted of the states of ‘employed’ and ‘unemployed’ and was run in yearly cycles.
People in the ‘employed’ state could remain in this state or move to the
‘unemployed’ state. Similarly, people in the ‘unemployed’ state could remain in this
state or move to the ‘employed’ state. In both arms of the Markov model, people
who were in the ‘unemployed’ state were assumed to receive TAU consisting of day
services for the duration of time they remained unemployed. It must be noted that
people in the ‘employed’ state were assumed to spend only a proportion of each year
in employment. A schematic diagram of the economic model is presented in Figure
10.
13.3.3 Costs and outcomes considered in the analysis

The economic analysis adopted the perspective of the NHS and PSS, as
recommended by NICE (2012c). The analysis considered intervention and TAU costs
and other NHS and PSS costs (including mental health, primary and secondary
care). The measure of outcome was the quality-adjusted life year (QALY). Clinical
input parameters of the economic model, including data on employment rates
following TAU and the relative effect of supported employment programmes versus
TAU at the end of the intervention period, were taken from the guideline systematic
review and meta-analysis that included three RCTs (CHANDLER1996, FREY2011,
KILLACKEY2008). Most of the published studies on supported employment report
outcomes at the end of the intervention, consequently less is known about vocational
outcomes over the long term.
Becker and colleagues (2007) conducted an exploratory study looking at 8 to 12-year

employment trajectories among adults with serious mental illness who participated
in the supported employment programme in a small urban mental health centre in
New England, USA. This was a follow-up study to two supported employment
research studies that were conducted at the same mental health centre in the early to
mid-1990s with 48 and 30 participants, respectively. No significant differences in
terms of patient characteristics were found between the two studies, therefore for the
long-term follow-up analysis participants from both studies were combined. The
authors could not contact 40 participants from the original two studies, therefore it
was assumed that all had lost their jobs. In total 38 participants were interviewed 8
to 12 years later and it was found that at the follow-up interview seven participants
worked 1 to 25% of time, four participants worked 26 to 50% of time, 14 participants
worked 51 to 75% and 13 participants worked 76 to 100% of time. Conservatively,

only those who worked for more that 50% of the follow-up time were considered
when estimating the probability of employment at 10 years’ follow-up. Based on the
above, the probability of employment at 10 years’ follow-up was estimated to be
0.35. Although the follow-up ranged from 8 to 12 years, the unemployment rate was
assumed to correspond to a mid-point of 10 years in order to estimate annual
probability of unemployment.

Figure 10: Schematic diagram of the structure of the economic model evaluating supported employment versus treatment as
usual (day services) for adults with psychosis and schizophrenia
Employed
Employed
Supported Unemployed
Employment
Unemployed
Adults with psychosis

Employed
and schizophrenia
Employed
Unemployed
Day services
Unemployed
22 months 10 years

Consequently, the annual transition probability of moving from the ‘employed’ to
the ‘unemployed’ health state over long-term follow-up in the model was estimated
to be 0.10. This rate was applied to both intervention and TAU groups, although it is
anticipated that people attending a supported employment programme are more
likely to retain their jobs after the end of the intervention compared with those under
TAU. If this is the case, then the economic analysis has underestimated the long-term
relative effect (in terms of remaining in paid employment) of supported employment
programmes versus TAU. The annual transition probability of moving from the
‘unemployed’ to the ‘employed’ health state over 10 years was estimated using data
from the studies included in the guideline systematic review (TAU arm). The same
rate was applied to both intervention and TAU groups. The mean time in
employment for every service user who remained in the ‘employed’ state of the
Markov model each year following completion of the intervention was derived from
the studies in the guideline systematic review—the average duration of employment
was 49% in the intervention group and 47% in the TAU group for every year of
employment. Clinical input parameters of the economic analysis are provided in
Table 164.
13.3.4 Utility data and estimation of QALYs

In order to express outcomes in the form of QALYs, the health states of the economic
model needed to be linked to appropriate utility scores. Utility scores represent the
health-related quality of life (HRQoL) associated with specific health states on a scale
from 0 (death) to 1 (perfect health); they are estimated using preference-based
measures that capture people’s preferences on the HRQoL experienced in the health
states under consideration.
The systematic search of the literature identified no studies reporting utility scores
for people with psychosis and schizophrenia. To estimate QALYs for adults with
psychosis and schizophrenia being in the two health states of ‘employed’ and
‘unemployed’, data reported in Squires and colleagues (2012), who conducted an
economic analysis to support the NICE public health guidance on managing long-
term sickness absence and incapacity for work (NICE, 2009b), were used. That
economic analysis (Squires et al., 2012) used utility scores for the health states of
‘being at work’ and ‘being on long-term sick leave’ estimated based on the findings
of a study aiming to predict the HRQoL of people who had been or were on long-
term sick leave (Peasgood et al., 2006), which utilised data from the British
Household Panel Survey (Taylor, 2003). This is a longitudinal annual survey
designed to capture information on a nationally representative sample of around
10,000 to 15,000 of the non-immigrant population of Great Britain that began in 1991.
Utility scores were estimated from Short Form Health Survey – 36-items (SF-36)
data, using the SF- 6D algorithm (Brazier et al., 2002). In the economic analysis
(Squires et al., 2012), the utility scores associated with being at work or being on
long-term sick leave were assumed to be the same for all individuals in each state,
independent of their health status; in other words, it was assumed that the quality of
life of the individual is more greatly affected by being at work or on sick leave than
by the illness itself. In addition, the utility scores for people at work and those on

sick leave were assumed to capture wage and benefit payments, respectively. Utility
scores were reported separately for four age categories (under 35 years; 35 to 45
years; 45 to 55 years; and over 55 years).
The economic analysis undertaken for this guideline used the utility scores reported
in Squires and colleagues (2012) for adults aged below 35 years, since the mean age
of participants in the studies included in the guideline systematic review ranged
from 21 to 47 years. Also, the difference in utility between the states of ‘being at
work’ and ‘being on sick leave’ was smaller in this age group (0.17) compared with
the 35 to 45 age group (0.21), thus providing a more conservative estimate and
potentially underestimating the benefit and the cost effectiveness of a supported
employment programme. It must be noted that the utility of the ‘unemployed’ state
is likely to be lower than the utility of ‘being on sick leave’, and therefore the
analysis is likely to have further underestimated the scope for benefit of a supported
employment programme. In addition, the utility scores used in the analysis refer to
the general population and are not specific to adults with psychosis and
schizophrenia. It is possible that adults with psychosis and schizophrenia get greater
utility from finding employment compared with the general population because
employment may bring them further benefits. Becker and colleagues (2007) reported
that there is evidence that increased employment has enduring benefits in terms of
better self-reported quality of life, self-esteem and relationships with other people.
Utility data used in the economic analysis are reported in Table 164.
13.3.5 Cost data

Cost data - intervention costs
Intervention costs for supported employment programmes and day care services
were based on Curtis (2012), who provided unit costs for IPS for four different
grades of staff: two with professional qualifications (for example, psychology or
occupational therapy) and two with no particular qualifications, ranging from Band
3 to Band 6, and for different caseloads, ranging from 10 to 25. Estimation of unit
costs for IPS took into account the following cost components: wages, salary on-
costs, superannuation, direct and indirect overheads, capital, team leaders who
would supervise no more than ten staff and would be available to provide practical
support, and a marketing budget. For this analysis, it was assumed that a supported
employment programme was provided by specialists in Band 6 with a caseload of 20
people. The average annual cost per person under these conditions was £3,594.
Curtis (2012) also provides unit costs for the equivalent of IPS in day care. In the
economic analysis, day care was conservatively assumed to be provided by
unqualified staff in Band 3, also with a caseload of 20 people. Curtis (2012) reported
that the number of day care sessions ranged from 34 to 131 annually. The lower
number of sessions (34) was selected for the economic analysis, resulting in an
annual cost of £1,938. All cost data input parameters are provided in Table 164.

Table 164: Input parameters utilised in the economic model of supported employment versus treatment as usual (day care
services) for adults with psychosis and schizophrenia
Input parameter Deterministic Probabilistic Source of data - comments

value distribution
Clinical input parameters
Probability of unemployment at 22 months – TAU 0.69 Beta distribution Guideline meta-analysis
α = 796, β = 362
Risk ratio of unemployment at 22 months– 0.46 Log-normal distribution Guideline meta-analysis
supported employment programme versus TAU 95% CI, 0.25 to 0.85
Probability of employment at 10 years’ follow-up 0.35 Beta distribution Becker et al. (2007); data on supported employment
α = 27, β = 51 utilised in both supported employment and treatment as
usual arms
Annual transition probability from ‘employed’ to 0.10 Distribution dependant -
‘unemployed’ on above distribution
Proportion of time employed with ‘employed state’ 0.47 Beta distribution Studies in the guideline meta-analysis
– standard care α = 9.43, β = 10.57
Proportion of time employed with ‘employed state’ 0.49 Beta distribution Studies in the guideline meta-analysis
– supported employment α = 9.77, β = 10.23
Utility scores Beta distribution Squires et al. (2012); utility scores for general population
Employed 0.83 α = 83, β = 17 being in work and on sick leave; distribution parameters
Unemployed 0.66 α = 66, β = 34 based on assumption
Cost data (2011/2012 prices)
Annual intervention cost Gamma distribution
Supported employment programme £3,594 α = 11.11, β = 323.46 Curtis (2012); standard error assumed to be 30% of its
TAU (day care services) £1,938 α = 11.11, β = 174.42 mean estimate because of lack of relevant data
Weekly health and social service cost Gamma distribution

Unemployed £47 α = 24.72, β = 1.92 Schneider et al. (2009); costs were up-rated to 2011/2012
Employed £36 α = 6.15, β = 5.85 prices using the pay and prices inflation index
Discount rate 0.035 N/A NICE (2012c)

It should be noted that the economic model utilised a 22-month cost for both
interventions for the initial period of provision. However, after entering the Markov
model, people in the ‘unemployed’ state were assumed to incur the annual cost of
day care services in every model cycle in which they remained unemployed, and this
applied to both arms of the model.
Cost data - NHS and PSS costs

Schneider and colleagues (2009) estimated the changes in costs to mental health,
primary and secondary care, local authority and voluntary day care services
incurred by people with mental health problems (mainly schizophrenia, bipolar
disorder, anxiety disorders or depression) associated with gaining employment
following registration with supported employment programmes.
The study reported baseline and 12-month follow-up data for people remaining
unemployed throughout the study (n = 77), people who found employment during
the 12 months between baseline and follow-up (n = 32), and people who were
already in employment at baseline and remained in employment at follow-up (n =
32). Cost data for people who found employment between baseline and follow-up
were utilised in the economic analysis; cost data at baseline were used for the state of
‘unemployed’; and cost data at follow-up were used for the state of ‘employed’ in
both the decision-tree and the Markov part of the model. Service costs included
mental health services (contacts with psychiatrist, psychologist, community
psychiatric nurse, attendance at a day centre, counselling or therapeutic group work,
and inpatient mental healthcare), primary care (contacts with GP, district nurse,
community physiotherapist, dentist or optician), local authority services (day centres
run by social services, home care and social work inputs), other secondary NHS care
(hospital outpatient appointments and inpatient care for needs other than mental
health) and a negligible amount of voluntary day care run by not-for-profit agencies
that are independent of the public sector (about 0.3 to 0.5% of the total cost).
Chandler and colleagues (1996) found greater decline in the number of service users
living in institutional settings over the 3-year period following registration with
supported employment programmes when compared with service users receiving
usual care. However, potential changes in accommodation type and related changes
in costs have not been considered in the economic analysis since such costs may
have already been included in local authority service costs reported by Schneider
and colleagues (2009) and there was a risk of double counting services. All costs
were expressed in 2012 prices, uplifted, where necessary, using the Hospital and
Community Health Services Pay and Prices Index (Curtis, 2012). Discounting of
costs and outcomes was undertaken at an annual rate of 3.5%, as recommended by
NICE (2012c).
13.3.6 Data analysis and presentation of the results

In order to take into account the uncertainty characterising the model input
parameters, a probabilistic analysis was undertaken, in which input parameters were

assigned probability distributions, rather than being expressed as point estimates
(Briggs et al., 2006b). Subsequently, 1000 iterations were performed, each drawing
random values out of the distributions fitted onto the model input parameters. Mean
costs and QALYs for each intervention were then calculated by averaging across
1000 iterations. The incremental cost-effectiveness ratio (ICER) was then estimated
expressing the additional cost per extra QALY gained associated with provision of
supported employment instead of TAU. The probability of employment for TAU
and the probability of employment at 10 years were given a beta distribution. Beta
distributions were also assigned to utility values and the proportion of time
employed within the ‘employed’ state. The risk ratio of supported employment
programmes versus TAU was assigned a log-normal distribution. Costs were
assigned a gamma distribution. The estimation of distribution ranges was based on
available data in the published sources of evidence, and further assumptions where
relevant data were not available. Table 164 provides details on the types of
distributions assigned to each input parameter and the methods employed to define
their range. Results of probabilistic analysis are also presented in the form of CEACs,
which demonstrate the probability of supported employment programmes being
cost effective relative to TAU at different levels of willingness-to-pay per QALY, that
is, at different cost-effectiveness thresholds the decision-maker may set (Fenwick et
al., 2001). One-way sensitivity analyses (run with the point estimates rather than the
distributions of the input parameters) explored the impact of the uncertainty
characterising the model input parameters on the model’s results: the intervention
cost for supported employment programmes and TAU was changed by ±50% to
investigate whether the conclusions of the analysis would change. In addition, a
threshold analysis explored the minimum relative effect of the supported
employment programme that is required in order for the intervention to be cost
effective using the NICE cost-effectiveness threshold.
Results
The results are presented in Table 165. Supported employment programmes are
associated with a higher cost but also produce a higher number of QALYs compared
with TAU. The ICER of supported employment programmes versus TAU is £5,723
per QALY gained, which is well below the NICE cost-effectiveness threshold of
£20,000 to £30,000 per QALY, indicating that supported employment programmes
may be a cost-effective option when compared with TAU. The cost effectiveness
plane showing the incremental costs and QALYs of supported employment
programmes versus TAU resulting from 1000 iterations of the model is shown in
Figure 11. According to the CEAC the probability of supported employment
programme being cost effective at the NICE lower cost-effectiveness threshold of
£20,000/QALY is 0.66, while at the NICE upper cost-effectiveness threshold of
£30,000/QALY it is 0.71.

Table 165: Results of economic analysis – mean total cost and QALYs of each
intervention at 10 years’ follow-up assessed per adult with psychosis and
schizophrenia seeking employment
Intervention Supported Treatment as Difference

employment usual
programmes
Total cost £34,239 £33,441 £798
Total QALYs 7.25 7.11 0.14
ICER £5,723/QALY
Figure 11: Cost effectiveness plane showing incremental costs and QALYs of
supported employment programme versus TAU (day care services) per adult with
psychosis or schizophrenia seeking employment. Results based on 1000 iterations.

One-way sensitivity analysis showed that as the risk ratio is varied across its range
the cost effectiveness of supported employment ranges from being dominant to
£48,307 per QALY gained. Also, threshold analysis revealed that the minimum risk
ratio of supported employment programmes versus TAU required in order for the
intervention to be considered cost effective according to NICE criteria was 0.69 using
the lower £20,000/QALY threshold and 0.77 using the upper £30,000/QALY
threshold. Moreover, as the intervention cost of supported employment programme
was changed by ±50%, the ICER ranged from £23,201/QALY to supported
employment being dominant and if the cost of TAU was changed by ±50%, then the
ICER ranged from a supported employment programme being dominant to £23,903
per QALY gained.
13.3.7 Discussion of findings – limitations of the analysis

The results of the economic analysis indicate that a supported employment
programme is likely to be a cost-effective intervention compared with TAU.
Supported employment programmes are associated with a higher cost but also
produce a higher number of QALYs compared with TAU. The ICER of supported
employment programmes versus TAU is £5,723 per QALY gained, which is well
below the NICE cost-effectiveness threshold of £20,000 to £30,000 per QALY. The
probability of supported employment programmes being cost effective at the NICE
lower cost-effectiveness threshold of £20,000/QALY was 0.66, while at the NICE
upper cost-effectiveness threshold it was 0.71.
In terms of clinical data, the economic analysis was based on three non-UK studies
comparing a supported employment programme with TAU. Frey and colleagues
(2011) conducted a large RCT (FREY2011) (n = 2,238) in service users with
schizophrenia spectrum or mood disorders across multiple locations in the USA.
Killackey and colleagues (2008) conducted a small RCT (KILLACKEY2008) (n = 41)
in service users with schizophrenia in Australia. Chandler and colleagues (1996)
undertook a medium-sized RCT (CHANDLER1996) (n = 256) in service users with
unspecified serious mental illness in the USA. It is not clear to what extent clinical
effectiveness can be generalised to the UK, given many structural differences in the
economy, the labour market, and health and social care systems between the USA,
Australia and the UK. Nevertheless, a recent review by Bond and colleagues (2012)
compared the results of nine RCTs of IPS in the USA with six RCTs outside the USA.
The authors examined competitive employment outcomes, including employment
rate, days to first job, weeks worked during follow-up, and hours worked. They also
considered non-competitive employment, programme retention and non-vocational
outcomes. It was found that the overall competitive employment rate for IPS clients
in US studies was significantly higher than in non-US studies (62% versus 47%).
However it was concluded that the consistently positive competitive employment
outcomes strongly favouring IPS over a range of comparison programmes in a group
of international studies suggest that IPS is an evidence-based practice that may
transport well into new settings as long as programmes achieve high fidelity to the
IPS model. In all studies included in the guideline meta-analysis the risk ratio of a

supported employment programme versus TAU in terms of vocational outcomes
was significant. The uncertainty in the clinical effectiveness estimate was assessed
using deterministic sensitivity analysis. It showed that as the risk ratio is varied
across its range the cost effectiveness of supported employment ranges from being
dominant to £48,307 per QALY gained, reflecting high uncertainty around the risk
ratio estimate. The threshold analysis revealed that the minimum risk ratio of
supported employment programmes versus TAU required in order for the
intervention to be considered cost effective according to NICE criteria was 0.69 using
the lower £20,000/QALY threshold and 0.77 using the upper £30,000/QALY
threshold.
In the studies used to assess the clinical effectiveness of supported employment

programmes in the guideline meta-analysis, TAU was defined as local mental health
services that included individual case management, medical review and other
rehabilitative services. A wide range of services provided under TAU and
inadequate information reported in the studies made it impossible to model TAU
according to these studies. According to the GDG, in the UK the current best
alternative to a supported employment programme would be a prevocational
training programme. However, given the lack of data pertaining to resource
utilisation associated with providing a prevocational training programme it was not
possible to cost it. Nevertheless, a prevocational programme is likely to be more
resource intensive than a supported employment programme as it is likely to
involve work crews, training, practising skills, job support, sheltered workshops,
and so on. Also, a greater mix of specialists is likely to be involved in providing a
prevocational programme including, but not limited to, mental health providers,
vocational counsellors, case managers, employment specialists, vocational staff, and
so on; usually prevocational programmes last longer because of the prolonged
preparation time. In the guideline systematic review it was found that more
participants gain competitive employment following a supported employment
programme compared with a prevocational programme (RR 0.63 [95% CI: 0.56;
0.72]). As a result, a supported employment programme is likely to be dominant
when compared with a prevocational training programme, that is, a supported
employment programme results in better clinical outcomes and lower costs.
Where data were not available or further estimates needed to be made, the economic
analysis always adopted conservative estimates that were likely to underestimate the
cost effectiveness of supported employment programmes. The intervention cost of
supported employment programme was estimated to be high because it was
assumed that the intervention was provided by specialists in Band 6. Given the lack
of data, in the economic analysis day care was defined as an alternative to a
supported employment programme. It was conservatively assumed to be provided
by unqualified staff in Band 3 and that the lower estimate of 34 annual sessions was
selected. The uncertainty associated with the definition of TAU and its associated
costs was assessed using deterministic sensitivity analysis. It was found that if the
cost of TAU was changed by as much as 50% the ICER ranged from a supported

employment programme being dominant to £23,903 per QALY gained, which is still
below the upper NICE cost-effectiveness threshold of £30,000 per QALY.
Also, most published RCT studies on supported employment report outcomes 12 to

24 months after first joining the programme. This is mainly because of the costs and
complexity of following up people for much longer periods of time, particularly
those who are no longer in receipt of services (Sainsbury Centre for Mental Health,
2009). Consequently, employment retention rates following a supported
employment programme were taken from an exploratory study looking at 8 to 12-
year employment trajectories among adults with serious mental illness who
participated in a supported employment programme. Becker and colleagues (2007)
interviewed 38 of 78 participants (49% with severe mental illness) 8 to 12 years after
they enrolled in supported employment studies in a small urban mental health
centre in New England, USA. This study reported that 35% of participants who
participated in supported employment programme were in employment during the
long-term follow-up which was used to estimate the annual probability of
employment. The same rate was applied to both intervention and TAU groups,
although service users attending a supported employment programme are more
likely to retain their jobs after the end of the intervention. If this was the case, then
the economic analysis has underestimated the long-term relative effects (in terms of
remaining in paid employment) of supported employment programme versus TAU.
Moreover, the rates were taken from a small USA-based study and it is questionable
how transferable the results are to the UK, given many structural differences in the
economy, labour market and health and welfare systems between the USA and other
countries (Sainsbury Centre for Mental Health, 2009). Regardless of the uncertainty
in the estimated employment retention rate the deterministic sensitivity analysis
indicated that even if it is assumed that as few as 5% of participants retained their
jobs at 10-year follow-up, the cost effectiveness of supported employment would be
£16,617 per QALY gained, which is still below the lower NICE cost-effectiveness
threshold of £20,000/QALY.
Moreover, the analysis considered extra NHS and PSS costs associated with
employment status. Cost data were taken from a small study (n = 77) by Schneider
and colleagues (2009), which measured costs incurred by people with mental health
problems including schizophrenia, bipolar disorder, anxiety disorders or depression
attending employment support programmes. The study reported that study
participants entering work showed a substantial decrease in mental health services
costs, which outweighed a slight increase in other secondary care costs, making an
overall reduction in health and social care costs statistically significant. The authors’
estimate was that the reduction in mental health service use was possibly an effect of
getting a job, although they did not rule out the possibility that a third variable, such
as cognitive impairment, might be driving both employment outcomes and
reduction in service use. The reported service costs within the analysis include those
that would typically fall on the NHS and PSS perspective, although some local
authority costs were also included such as day centres run by social services, home
care and other social work inputs. The local authority costs accounted for

approximately 10% of service costs; the deterministic sensitivity analysis indicated
that reducing service costs by 10% resulted in a cost per QALY of £6,794, which is
still well below the lower NICE cost-effectiveness threshold of £20,000/QALY. Also,
according to the GDG, some of the aforementioned services could be provided by a
range of providers including the NHS and PSS. Some trusts (but not all) provide
social care/social work input on behalf of the local authorities, consequently some of
the local authority costs may be relevant from the NHS and PSS perspective.
Utility scores, which are required for the estimation of QALYs, were not available for
adults with psychosis and schizophrenia. Instead, utility scores obtained from the
general population for the states ‘being at work’ and ‘being on sick leave’ were used
in the analysis, based on data reported in Squires and colleagues (2012). It is
acknowledged that these scores are not directly relevant to adults with psychosis
and schizophrenia in employed or unemployed status. Moreover, the utility of the
‘unemployed’ state is potentially lower than the utility of ‘being on sick leave’.
Nevertheless, the utility scores used in the economic analysis are likely to capture, if
somewhat conservatively, the HRQoL of adults with psychosis and schizophrenia
with regard to their employment status. Also it is possible that adults with severe
mental illness may get greater utility from finding employment compared with the
general population, as employment may bring further psychological and social
benefits, including enhancements to self-esteem, relationships and illness
management (Becker et al., 2007).
The analysis adopted the NHS and PSS perspective. Other costs, such as lost
productivity or wages earned and the tax gains to the exchequer, and reduction in
welfare benefits, were not taken into account because they were beyond the
perspective of the analysis. Also such programmes have a positive effect on the
HRQoL of families, partners and carers of adults with psychosis and schizophrenia,
which was not possible to capture in the economic analysis.
13.3.8 Validation of the economic model

The economic model (including the conceptual model and the Excel spread sheet)
was developed by the guideline health economist and checked by a second modeller
not working on the guideline. The model was tested for logical consistency by
setting input parameters to null and extreme values and examining whether results
changed in the expected direction. The results were discussed with the GDG for their
plausibility.
13.3.9 Overall conclusions from economic modelling

Overall, although based on limited evidence, the findings of the economic analysis
indicate that a supported employment programme is potentially a cost-effective
intervention for adults with psychosis and schizophrenia because it can increase the
rate of employment in this population group, improve the person’s wellbeing, and
potentially reduce the economic burden to health and social services and the wider
society.

The GDG agreed that the main aim of a vocational rehabilitation intervention is to
get people into employment and to improve functioning and quality of life. For
cognitive remediation with vocational rehabilitation, the aim of the review was to
evaluate if the addition of a cognitive remediation intervention to vocational
rehabilitation improved vocational outcomes and not if they improved cognitive
outcomes (the efficacy of cognitive remediation alone is evaluated in Chapter 9).
Therefore, the GDG judged that employment and education, quality of life and
functional disability were critical outcomes. Important, but not critical, outcomes
were considered to be adverse effects, effects on symptom-focused outcomes and
service use, as well as satisfaction with services and acceptability. Although these
outcomes were not considered critical in informing recommendations for the
benefits of vocational rehabilitation on the outcomes pertinent to the intervention
(vocational and functioning), they informed the GDG about the feasibility of the
intervention.
Trade-off between clinical benefits and harms:

For adults with psychosis and schizophrenia, the GDG considered there to be
reasonable evidence that the benefits of a supported employment intervention
outweigh the possible risk of harm (for example, relapse due to the negative effects
of being employed). The evidence suggests that vocational rehabilitation (all
formats) is more effective than a non-vocational intervention/control for gaining
employment (competitive or otherwise) and although any additional benefit on
functioning or quality of life is uncertain and varied across interventions, it also does
not adversely affect psychological health or exacerbate psychotic symptoms.
Furthermore, supported employment was more effective than prevocational training
for vocational outcomes and equal to prevocational training for functioning and
quality of life outcomes, and did not have a harmful effect on psychological health
(for example, hospital admissions and psychological distress).
The GDG felt there was a paucity of follow-up data evaluating the long-term efficacy
of vocational rehabilitation interventions. However, the group believed that the
potential negative consequences of not being offered any vocational support
outweighed the lack of confidence in the long-term benefits.

For adults with psychosis and schizophrenia the health economic evidence for
supported employment versus prevocational training is limited to one UK-based
study. The GDG felt that prevocational training is likely to be more resource
intensive and is expected to be more expensive than supported employment
intervention. The international evidence is mixed. One study undertaken across six
European sites found IPS dominant when compared with standard care in all but
one site. However, the study undertaken in the USA could not reach firm

conclusions pertaining to the cost effectiveness of IPS. According to the guideline
economic analysis, for adults with psychosis and schizophrenia a supported
employment intervention appears to be cost effective when compared with a non-
vocational intervention or control. Despite limitations in the economic analysis (for
instance, weak and mainly USA-based evidence for the clinical effectiveness, lack of
long-term follow-up data, lack of data pertaining to treatment as usual, utility values
specific for this population not being available), the findings were robust to
underlying assumptions. In general, the health economic evidence supports the
GDG’s view that a vocational rehabilitation intervention should be provided.

For supported employment versus prevocational training, the evidence ranged from
very low to high. Reasons for downgrading concerned risk of bias, high
heterogeneity or lack of precision in confidence intervals. Heterogeneity was a major
concern when evaluating the evidence. The interventions and controls varied
between studies. However, although variance was observed in the effect size across
studies, the direction of effect was consistent across most studies.
The evidence suggested that any vocational rehabilitation intervention was
beneficial on quality of life and functioning outcomes compared with a non-
vocational control group. The GDG felt that this finding supported their
recommendation that a vocational rehabilitation intervention should be provided.
The evidence also suggested that supported employment is more effective than
prevocational training for gaining competitive employment. The GDG judged that
this would only be appropriate for those who desired competitive employment. For
those who need a more gradual introduction into work and would like support
before entering into competitive employment, there is some evidence of efficacy for
prevocational training. The GDG believed that there should be an element of choice
for the service user, with those seeking immediate competitive employment to have
the option of supported employment, and those unable to return to work
immediately being provided with support and training before attempting to gain
competitive employment. The GDG discussed collaboration between various local
stakeholders to ensure the service user is supported in education, and obtaining and
retaining occupation and employment. It was decided that this should include local
stakeholders for black, Asian and minority ethnic groups. The GDG also discussed
that vocational employment, education, or any daytime activities should be
monitored and a part of the care plan.
The majority of the evidence base was from the USA and sub-analyses revealed that
the benefit of vocational rehabilitation interventions was not as compelling in studies
based in only the UK or Europe, although the same trends were observed. Although
the GDG felt this was of some concern, it highlights the need for more trials
evaluating services provided in the UK.

The evidence base for the combined intervention of cognitive remediation and
vocational rehabilitation was found to be too limited to make a recommendation and
the GDG identified this as potential topic for a research recommendation for more
UK-based studies.
13.5 RECOMMENDATIONS
13.5.1.1 For people who are unable to attend mainstream education, training or
work, facilitate alternative educational or occupational activities according
to their individual needs and capacity to engage with such activities, with an
ultimate goal of returning to mainstream education, training or
employment. [new 2014]
13.5.1.2 Offer supported employment programmes to people with psychosis or
schizophrenia who wish to find or return to work. Consider other
occupational or educational activities, including pre-vocational training, for
people who are unable to work or unsuccessful in finding employment.
[new 2014]
13.5.1.3 Mental health services should work in partnership with local stakeholders,
including those representing black, Asian and minority ethnic groups, to
enable people with mental health problems, including psychosis or
schizophrenia, to stay in work or education and to access new employment
(including self-employment), volunteering and educational opportunities.
[2009; amended 2014]
13.5.1.4 Routinely record the daytime activities of people with psychosis or
schizophrenia in their care plans, including occupational outcomes. [2009]

14 SUMMARY OF
RECOMMENDATIONS
14.1 CARE ACROSS ALL PHASES
14.1.1 Service user experience
14.1.1.1 Use this guideline in conjunction with Service user experience in adult
mental health (NICE clinical guidance 136) to improve the experience of care
for people with psychosis or schizophrenia using mental health services,
and:
• work in partnership with people with schizophrenia and their
carers
• offer help, treatment and care in an atmosphere of hope and
optimism
• take time to build supportive and empathic relationships as an
essential part of care. [2009; amended 2014]
14.1.2 Race, culture and ethnicity

The NICE guideline on service user experience in adult mental health (NICE clinical
guidance 136) includes recommendations on communication relevant to this section.
14.1.2.1 Healthcare professionals inexperienced in working with people with
psychosis or schizophrenia from diverse ethnic and cultural backgrounds
should seek advice and supervision from healthcare professionals who are
experienced in working transculturally. [2009]
14.1.2.2 Healthcare professionals working with people with psychosis or
schizophrenia should ensure they are competent in:
• assessment skills for people from diverse ethnic and cultural
backgrounds
• using explanatory models of illness for people from diverse ethnic
and cultural backgrounds
• explaining the causes of psychosis or schizophrenia and treatment
options
• addressing cultural and ethnic differences in treatment
expectations and adherence
• addressing cultural and ethnic differences in beliefs regarding
biological, social and family influences on the causes of abnormal
mental states
• negotiating skills for working with families of people with
psychosis or schizophrenia
• conflict management and conflict resolution. [2009]

14.1.2.3 Mental health services should work with local voluntary black, Asian and
minority ethnic groups to jointly ensure that culturally appropriate
psychological and psychosocial treatment, consistent with this guideline and
delivered by competent practitioners, is provided to people from diverse
ethnic and cultural backgrounds. [2009]
14.1.3 Physical health

14.1.3.1 People with psychosis or schizophrenia, especially those taking
antipsychotics, should be offered a combined healthy eating and physical
activity programme by their mental healthcare provider. [new 2014]
14.1.3.2 If a person has rapid or excessive weight gain, abnormal lipid levels or
problems with blood glucose management, offer interventions in line with
relevant NICE guidance (see Obesity [NICE clinical guideline 43], Lipid
modification [NICE clinical guideline 67] and Preventing type 2 diabetes
[NICE public health guidance 38]). [new 2014]
14.1.3.3 Offer people with psychosis or schizophrenia who smoke help to stop
smoking, even if previous attempts have been unsuccessful. Be aware of the
potential significant impact of reducing cigarette smoking on the metabolism
of other drugs, particularly clozapine and olanzapine. [new 2014]
14.1.3.4 Consider one of the following to help people stop smoking:
•nicotine replacement therapy (usually a combination of
transdermal patches with a short-acting product such as an
inhalator, gum, lozenges or spray) for people with psychosis or
schizophrenia or
• bupropion 58 for people with a diagnosis of schizophrenia or
• varenicline for people with psychosis or schizophrenia.
Warn people taking bupropion or varenicline that there is an increased risk of
adverse neuropsychiatric symptoms and monitor them regularly, particularly in the
first 2–3 weeks. [new 2014]
14.1.3.5 For people in inpatient settings who do not want to stop smoking, offer
nicotine replacement therapy to help them to reduce or temporarily stop
smoking. [new 2014]
14.1.3.6 Routinely monitor weight, and cardiovascular and metabolic indicators of
morbidity in people with psychosis and schizophrenia. These should be
audited in the annual team report. [new 2014]
14.1.3.7 Trusts should ensure compliance with quality standards on the monitoring
and treatment of cardiovascular and metabolic disease in people with
psychosis or schizophrenia through board-level performance indicators.
[new 2014]
58 At the time of publication (February 2014), bupropion was contraindicated in people with bipolar disorder.
Therefore, it is not recommended for people with psychosis unless they have a diagnosis of schizophrenia.

14.1.4 Comprehensive services provision
14.1.4.1 All teams providing services for people with psychosis or schizophrenia
should offer a comprehensive range of interventions consistent with this
guideline. [2009]
14.1.5 Support for carers

14.1.5.1 Offer carers of people with psychosis or schizophrenia an assessment
(provided by mental health services) of their own needs and discuss with
them their strengths and views. Develop a care plan to address any
identified needs, give a copy to the carer and their GP and ensure it is
reviewed annually. [new 2014]
14.1.5.2 Advise carers about their statutory right to a formal carer’s assessment
provided by social care services and explain how to access this. [new 2014]
14.1.5.3 Give carers written and verbal information in an accessible format about:
• diagnosis and management of psychosis and schizophrenia
• positive outcomes and recovery
• types of support for carers
• role of teams and services
• getting help in a crisis.
When providing information, offer the carer support if necessary. [new 2014]
14.1.5.4 As early as possible negotiate with service users and carers about how
information about the service user will be shared. When discussing rights to
confidentiality, emphasise the importance of sharing information about risks
and the need for carers to understand the service user’s perspective. Foster a
collaborative approach that supports both service users and carers, and
respects their individual needs and interdependence. [new 2014]
14.1.5.5 Review regularly how information is shared, especially if there are
communication and collaboration difficulties between the service user and
carer. [new 2014]
14.1.5.6 Include carers in decision-making if the service user agrees. [new 2014]
14.1.5.7 Offer a carer-focused education and support programme, which may be part
of a family intervention for psychosis and schizophrenia, as early as possible
to all carers. The intervention should:
• be available as needed
• have a positive message about recovery. [new 2014]

14.1.6 Peer support and self-management
14.1.6.1 Consider peer support for people with psychosis or schizophrenia to help
improve service user experience and quality of life. Peer support should be
delivered by a trained peer support worker who has recovered from
psychosis or schizophrenia and remains stable. Peer support workers should
receive support from their whole team, and support and mentorship from
experienced peer workers. [new 2014]
14.1.6.2 Consider a manualised self-management programme delivered face-to-face
with service users, as part of the treatment and management of psychosis or
schizophrenia. [new 2014]
14.1.6.3 Peer support and self-management programmes should include information
and advice about:
• psychosis and schizophrenia
• effective use of medication
• identifying and managing symptoms
• accessing mental health and other support services
• coping with stress and other problems
• what to do in a crisis
• building a social support network
• preventing relapse and setting personal recovery goals. [new 2014]
14.2 PREVENTING PSYCHOSIS

14.2.1 Referral from primary care
14.2.1.1 If a person is distressed, has a decline in social functioning and has:
• transient or attenuated psychotic symptoms or
• other experiences or behaviour suggestive of possible psychosis or
• a first-degree relative with psychosis or schizophrenia
refer them for assessment without delay to a specialist mental health service
or an early intervention in psychosis service because they may be at increased
risk of developing psychosis. [new 2014]
14.2.2 Specialist assessment

14.2.2.1 A consultant psychiatrist or a trained specialist with experience in at-risk
mental states should carry out the assessment. [new 2014]
14.2.3 Treatment options to prevent psychosis

14.2.3.1 If a person is considered to be at increased risk of developing psychosis (as
described in recommendation 14.2.1.1):
• offer individual cognitive behavioural therapy (CBT) with or
without family intervention (delivered as described in
recommendations 14.3.7.1 and 14.3.7.2) and

• offer interventions recommended in NICE guidance for people
with any of the anxiety disorders, depression, emerging
personality disorder or substance misuse. [new 2014]
14.2.3.2 Do not offer antipsychotic medication:
• to people considered to be at increased risk of developing
psychosis (as described in recommendation 14.2.1.1) or
• with the aim of decreasing the risk of or preventing psychosis.
[new 2014]
14.2.4 Monitoring and follow-up

14.2.4.1 If, after treatment (as described in recommendation 14.2.3.1), the person
continues to have symptoms, impaired functioning or is distressed, but a
clear diagnosis of psychosis cannot be made, monitor the person regularly
for changes in symptoms and functioning for up to 3 years using a
structured and validated assessment tool. Determine the frequency and
duration of monitoring by the:
• severity and frequency of symptoms
• level of impairment and/or distress and
• degree of family disruption or concern. [new 2014]
14.2.4.2 If a person asks to be discharged from the service, offer follow-up
appointments and the option to self-refer in the future. Ask the person’s GP
to continue monitoring changes in their mental state. [new 2014]
14.3 FIRST EPISODE PSYCHOSIS

14.3.1 Early intervention in psychosis services
14.3.1.1 Early intervention in psychosis services should be accessible to all people
with a first episode or first presentation of psychosis, irrespective of the
person’s age or the duration of untreated psychosis. [new 2014]
14.3.1.2 People presenting to early intervention in psychosis services should be
assessed without delay. If the service cannot provide urgent intervention for
people in a crisis, refer the person to a crisis resolution and home treatment
team (with support from early intervention in psychosis services). Referral
may be from primary or secondary care (including other community
services) or a self- or carer-referral. [new 2014]
14.3.1.3 Early intervention in psychosis services should aim to provide a full range of
pharmacological, psychological, social, occupational and educational
interventions for people with psychosis, consistent with this guideline.
[2014]
14.3.1.4 Consider extending the availability of early intervention in psychosis
services beyond 3 years if the person has not made a stable recovery from
psychosis or schizophrenia. [new 2014]

14.3.2 Primary care
14.3.2.1 Do not start antipsychotic medication for a first presentation of sustained
psychotic symptoms in primary care unless it is done in consultation with a
consultant psychiatrist. [2009; amended 2014]
14.3.3 Assessment and care planning

14.3.3.1 Carry out a comprehensive multidisciplinary assessment of people with
psychotic symptoms in secondary care. This should include assessment by a
psychiatrist, a psychologist or a professional with expertise in the
psychological treatment of people with psychosis or schizophrenia. The
assessment should address the following domains:
• psychiatric (mental health problems, risk of harm to self or others,
alcohol consumption and prescribed and non-prescribed drug
history)
• medical, including medical history and full physical examination to
identify physical illness (including organic brain disorders) and
prescribed drug treatments that may result in psychosis
• physical health and wellbeing (including weight, smoking,
nutrition, physical activity and sexual health)
• psychological and psychosocial, including social networks,
relationships and history of trauma
• developmental (social, cognitive and motor development and
skills, including coexisting neurodevelopmental conditions)
• social (accommodation, culture and ethnicity, leisure activities and
recreation, and responsibilities for children or as a carer)
• occupational and educational (attendance at college, educational
attainment, employment and activities of daily living)
• quality of life
• economic status. [2009; amended 2014]
14.3.3.2 Assess for post-traumatic stress disorder and other reactions to trauma
because people with psychosis or schizophrenia are likely to have
experienced previous adverse events or trauma associated with the
development of the psychosis or as a result of the psychosis itself. For people
who show signs of post-traumatic stress, follow the recommendations in
Post-traumatic stress disorder (NICE clinical guideline 26). [new 2014]
14.3.3.3 Routinely monitor for other coexisting conditions, including depression,
anxiety and substance misuse particularly in the early phases of treatment.
[2009; amended 2014]
14.3.3.4 Write a care plan in collaboration with the service user as soon as possible
following assessment, based on a psychiatric and psychological formulation,
and a full assessment of their physical health. Send a copy of the care plan to
the primary healthcare professional who made the referral and the service
user. [2009; amended 2014]

14.3.3.5 For people who are unable to attend mainstream education, training or
work, facilitate alternative educational or occupational activities according
to their individual needs and capacity to engage with such activities, with an
ultimate goal of returning to mainstream education, training or
employment. [new 2014]
14.3.4 Treatment options

• oral antipsychotic medication (see sections 14.3.5.and 14.3.6) in
conjunction with
• psychological interventions (family intervention and individual
CBT, delivered as described in recommendations 14.3.7.1 and
14.3.7.2). [new 2014]
14.3.4.2 Advise people who want to try psychological interventions alone that these
are more effective when delivered in conjunction with antipsychotic
medication. If the person still wants to try psychological interventions alone:
• offer family intervention and CBT
• agree a time (1 month or less) to review treatment options,
including introducing antipsychotic medication
• continue to monitor symptoms, distress, impairment and level of
functioning (including education, training and employment)
regularly. [new 2014]
14.3.4.3 If the person’s symptoms and behaviour suggest an affective psychosis or
disorder, including bipolar disorder and unipolar psychotic depression,
follow the recommendations in Bipolar disorder (NICE clinical guideline 38)
or Depression (NICE clinical guideline 90). [new 2014]
14.3.5 Choice of antipsychotic medication

14.3.5.1 The choice of antipsychotic medication should be made by the service user
and healthcare professional together, taking into account the views of the
carer if the service user agrees. Provide information and discuss the likely
benefits and possible side effects of each drug, including:
• metabolic (including weight gain and diabetes)
• extrapyramidal (including akathisia, dyskinesia and dystonia)
• cardiovascular (including prolonging the QT interval)
• hormonal (including increasing plasma prolactin)
• other (including unpleasant subjective experiences). [2009;
amended 2014]
14.3.6 How to use antipsychotic medication

14.3.6.1 Before starting antipsychotic medication, undertake and record the
following baseline investigations:

• weight (plotted on a chart)
• waist circumference
• pulse and blood pressure
• fasting blood glucose, glycosylated haemoglobin (HbA1c), blood
lipid profile and prolactin levels
• assessment of any movement disorders
• assessment of nutritional status, diet and level of physical activity.
[new 2014]
14.3.6.2 Before starting antipsychotic medication, offer the person with psychosis or
schizophrenia an electrocardiogram (ECG) if:
• specified in the summary of product characteristics (SPC)
• a physical examination has identified specific cardiovascular risk
(such as diagnosis of high blood pressure)
• there is a personal history of cardiovascular disease or
• the service user is being admitted as an inpatient. [2009]
14.3.6.3 Treatment with antipsychotic medication should be considered an explicit
individual therapeutic trial. Include the following:
• Discuss and record the side effects that the person is most willing
to tolerate.
• Record the indications and expected benefits and risks of oral
antipsychotic medication, and the expected time for a change in
symptoms and appearance of side effects.
• At the start of treatment give a dose at the lower end of the licensed
range and slowly titrate upwards within the dose range given in
the British national formulary (BNF) or SPC.
• Justify and record reasons for dosages outside the range given in
the BNF or SPC.
• Record the rationale for continuing, changing or stopping
medication, and the effects of such changes.
• Carry out a trial of the medication at optimum dosage for 4–6
weeks. [2009; amended 2014]
14.3.6.4 Monitor and record the following regularly and systematically throughout
treatment, but especially during titration:
• response to treatment, including changes in symptoms and
behaviour
• side effects of treatment, taking into account overlap between
certain side effects and clinical features of schizophrenia (for
example, the overlap between akathisia and agitation or anxiety)
and impact on functioning
• the emergence of movement disorders
• weight, weekly for the first 6 weeks, then at 12 weeks, at 1 year and
then annually (plotted on a chart)
• waist circumference annually (plotted on a chart)

• pulse and blood pressure at 12 weeks, at 1 year and then annually
• fasting blood glucose, HbA1c and blood lipid levels at 12 weeks, at
1 year and then annually
• adherence
• overall physical health. [new 2014]
14.3.6.5 The secondary care team should maintain responsibility for monitoring
service users’ physical health and the effects of antipsychotic medication for
at least the first 12 months or until the person’s condition has stabilised,
whichever is longer. Thereafter, the responsibility for this monitoring may
be transferred to primary care under shared care arrangements. [new 2014]
14.3.6.6 Discuss any non-prescribed therapies the service user wishes to use
(including complementary therapies) with the service user, and carer if
appropriate. Discuss the safety and efficacy of the therapies, and possible
interference with the therapeutic effects of prescribed medication and
psychological treatments. [2009]
14.3.6.7 Discuss the use of alcohol, tobacco, prescription and non-prescription
medication and illicit drugs with the service user, and carer if appropriate.
Discuss their possible interference with the therapeutic effects of prescribed
medication and psychological treatments. [2009]
14.3.6.8 ‘As required’ (p.r.n.) prescriptions of antipsychotic medication should be
made as described in recommendation 14.3.6.3. Review clinical indications,
frequency of administration, therapeutic benefits and side effects each week
or as appropriate. Check whether ‘p.r.n.’ prescriptions have led to a dosage
above the maximum specified in the BNF or SPC. [2009]
14.3.6.9 Do not use a loading dose of antipsychotic medication (often referred to as
‘rapid neuroleptisation’). [2009]
14.3.6.10 Do not initiate regular combined antipsychotic medication, except for
short periods (for example, when changing medication). [2009]
14.3.6.11 If prescribing chlorpromazine, warn of its potential to cause skin
photosensitivity. Advise using sunscreen if necessary. [2009]
14.3.7 How to deliver psychological interventions

14.3.7.1 CBT should be delivered on a one-to-one basis over at least 16 planned
session and:
• Follow a treatment manual 59 so that:
- people can establish links between their thoughts, feelings or
actions and their current or past symptoms, and/or functioning
- the re-evaluation of people’s perceptions, beliefs or reasoning
relates to the target symptoms
59 Treatment manuals that have evidence for their efficacy from clinical trials are preferred.

• also include at least one of the following components:
- people monitoring their own thoughts, feelings or behaviours with
respect to their symptoms or recurrence of symptoms
- promoting alternative ways of coping with the target symptom
- reducing distress
- improving functioning. [2009]
14.3.7.2 Family intervention should:
• include the person with psychosis or schizophrenia if practical
• be carried out for between 3 months and 1 year
• include at least 10 planned sessions
• take account of the whole family's preference for either single-
family intervention or multi-family group intervention
• take account of the relationship between the main carer and the
person with psychosis or schizophrenia
• have a specific supportive, educational or treatment function and
include negotiated problem solving or crisis management work.
[2009]
14.3.8 Monitoring and reviewing psychological interventions

14.3.8.1 When providing psychological interventions, routinely and systematically
monitor a range of outcomes across relevant areas, including service user
satisfaction and, if appropriate, carer satisfaction. [2009]
14.3.8.2 Healthcare teams working with people with psychosis or schizophrenia
should identify a lead healthcare professional within the team whose
responsibility is to monitor and review:
• access to and engagement with psychological interventions
• decisions to offer psychological interventions and equality of access
across different ethnic groups. [2009]
14.3.9 Competencies for delivering psychological interventions

14.3.9.1 Healthcare professionals providing psychological interventions should:
• have an appropriate level of competence in delivering the
intervention to people with psychosis or schizophrenia
• be regularly supervised during psychological therapy by a
competent therapist and supervisor. [2009]
14.3.9.2 Trusts should provide access to training that equips healthcare professionals
with the competencies required to deliver the psychological therapy
interventions recommended in this guideline. [2009]

14.4 SUBSEQUENT ACUTE EPISODES OF PSYCHOSIS OR
SCHIZOPHRENIA AND REFERRAL IN CRISIS
14.4.1 Service-level interventions
14.4.1.1 Offer crisis resolution and home treatment teams as a first-line service to
support people with psychosis or schizophrenia during an acute episode in
the community if the severity of the episode, or the level of risk to self or
others, exceeds the capacity of the early intervention in psychosis services or
other community teams to effectively manage it. [new 2014]
14.4.1.2 Crisis resolution and home treatment teams should be the single point of
entry to all other acute services in the community and in hospitals. [new
2014]
14.4.1.3 Consider acute community treatment within crisis resolution and home
treatment teams before admission to an inpatient unit and as a means to
enable timely discharge from inpatient units. Crisis houses or acute day
facilities may be considered in addition to crisis resolution and home
treatment teams depending on the person’s preference and need. [new
2014]
14.4.1.4 If a person with psychosis or schizophrenia needs hospital care, think about
the impact on the person, their carers and other family members, especially
if the inpatient unit is a long way from where they live. If hospital admission
is unavoidable, ensure that the setting is suitable for the person’s age,
gender and level of vulnerability, support their carers and follow the
recommendations in Service user experience in adult mental health (NICE
14.4.2 Treatment options

• oral antipsychotic medication in conjunction (see sections 14.3.5.
and 14.3.6 with
• psychological interventions (family intervention and individual
CBT, delivered as described in recommendations 14.3.7.1 and
14.3.7.2). [new 2014]
14.4.3 Pharmacological interventions

schizophrenia, offer oral antipsychotic medication or review existing
medication. The choice of drug should be influenced by the same criteria
recommended for starting treatment (see sections 14.3.5.and 14.3.6). Take
into account the clinical response and side effects of the service user’s
current and previous medication. [2009; amended 2014]

14.4.4 Psychological and psychosocial interventions
14.4.4.1 Offer CBT to all people with psychosis or schizophrenia (delivered as
described in recommendation 14.3.7.1). This can be started either during the
acute phase or later, including in inpatient settings. [2009]
14.4.4.2 Offer family intervention to all families of people with psychosis or
schizophrenia who live with or are in close contact with the service user
(delivered as described in recommendation 14.3.7.2). This can be started
either during the acute phase or later, including in inpatient settings. [2009]
14.4.4.3 Consider offering arts therapies to all people with psychosis or
schizophrenia, particularly for the alleviation of negative symptoms. This
can be started either during the acute phase or later, including in inpatient
settings. [2009]
14.4.4.4 Arts therapies should be provided by a Health and Care Professions Council
registered arts therapist with previous experience of working with people
with psychosis or schizophrenia. The intervention should be provided in
groups unless difficulties with acceptability and access and engagement
indicate otherwise. Arts therapies should combine psychotherapeutic
techniques with activity aimed at promoting creative expression, which is
often unstructured and led by the service user. Aims of arts therapies should
include:
• enabling people with psychosis or schizophrenia to experience
themselves differently and to develop new ways of relating to
others
• helping people to express themselves and to organise their
experience into a satisfying aesthetic form
• helping people to accept and understand feelings that may have
emerged during the creative process (including, in some cases, how
they came to have these feelings) at a pace suited to the person.
[2009]
14.4.4.5 When psychological treatments, including arts therapies, are started in the
acute phase (including in inpatient settings), the full course should be
continued after discharge without unnecessary interruption. [2009]
14.4.4.6 Do not routinely offer counselling and supportive psychotherapy (as specific
interventions) to people with psychosis or schizophrenia. However, take
service user preferences into account, especially if other more efficacious
psychological treatments, such as CBT, family intervention and arts
therapies, are not available locally. [2009]
14.4.4.7 Do not offer adherence therapy (as a specific intervention) to people with
psychosis or schizophrenia. [2009]
14.4.4.8 Do not routinely offer social skills training (as a specific intervention) to
people with psychosis or schizophrenia. [2009]

14.4.5 Behaviour that challenges
14.4.5.1 Occasionally people with psychosis or schizophrenia pose an immediate risk
to themselves or others during an acute episode and may need rapid
tranquillisation. The management of immediate risk should follow the
relevant NICE guidelines (see recommendations 14.4.5.2 and 14.4.5.5). [2009]
14.4.5.2 Follow the recommendations in Violence (NICE clinical guideline 25) when
facing imminent violence or when considering rapid tranquillisation. [2009]
14.4.5.3 After rapid tranquillisation, offer the person with psychosis or schizophrenia
the opportunity to discuss their experiences. Provide them with a clear
explanation of the decision to use urgent sedation. Record this in their notes.
[2009]
14.4.5.4 Ensure that the person with psychosis or schizophrenia has the opportunity
to write an account of their experience of rapid tranquillisation in their
notes. [2009]
14.4.5.5 Follow the recommendations in Self-harm (NICE clinical guideline 16) when
managing acts of self-harm in people with psychosis or schizophrenia. [2009]
14.4.6 Early post-acute period

14.4.6.1 After each acute episode, encourage people with psychosis or schizophrenia
to write an account of their illness in their notes. [2009]
14.4.6.2 Healthcare professionals may consider using psychoanalytic and
psychodynamic principles to help them understand the experiences of
people with psychosis or schizophrenia and their interpersonal
relationships. [2009]
14.4.6.3 Inform the service user that there is a high risk of relapse if they stop
medication in the next 1–2 years. [2009]
14.4.6.4 If withdrawing antipsychotic medication, undertake gradually and monitor
regularly for signs and symptoms of relapse. [2009]
14.4.6.5 After withdrawal from antipsychotic medication, continue monitoring for
signs and symptoms of relapse for at least 2 years. [2009]
14.5 PROMOTING RECOVERY AND POSSIBLE FUTURE

CARE
14.5.1 General principles
14.5.1.1 Continue treatment and care in early intervention in psychosis services or
refer the person to a specialist integrated community-based team. This team
should:
• offer the full range of psychological, pharmacological, social and
occupational interventions recommended in this guideline

• be competent to provide all interventions offered
• place emphasis on engagement rather than risk management
• provide treatment and care in the least restrictive and stigmatising
environment possible and in an atmosphere of hope and optimism
in line with Service user experience in adult mental health (NICE
14.5.1.2 Consider intensive case management for people with psychosis or
schizophrenia who are likely to disengage from treatment or services. [new
2014]
14.5.1.3 Review antipsychotic medication annually, including observed benefits and
any side effects. [new 2014].
14.5.2 Return to primary care

14.5.2.1 Offer people with psychosis or schizophrenia whose symptoms have
responded effectively to treatment and remain stable the option to return to
primary care for further management. If a service user wishes to do this,
record this in their notes and coordinate transfer of responsibilities through
the care programme approach. [2009]
14.5.3 Primary care

Monitoring physical health in primary care
14.5.3.1 Develop and use practice case registers to monitor the physical and mental
health of people with psychosis or schizophrenia in primary care. [2009]
14.5.3.2 GPs and other primary healthcare professionals should monitor the physical
health of people with psychosis or schizophrenia when responsibility for
monitoring is transferred from secondary care, and then at least annually.
The health check should be comprehensive, focusing on physical health
problems that are common in people with psychosis and schizophrenia.
Include all the checks recommended in 14.3.6.1and refer to relevant NICE
guidance on monitoring for cardiovascular disease, diabetes, obesity and
respiratory disease. A copy of the results should be sent to the care
coordinator and psychiatrist, and put in the secondary care notes. [new
2014]
14.5.3.3 Identify people with psychosis or schizophrenia who have high blood
pressure, have abnormal lipid levels, are obese or at risk of obesity, have
diabetes or are at risk of diabetes (as indicated by abnormal blood glucose
levels), or are physically inactive, at the earliest opportunity following
relevant NICE guidance (see Lipid modification [NICE clinical guideline 67],
Preventing type 2 diabetes [NICE public health guidance 38], Obesity [NICE
clinical guideline 43], Hypertension [NICE clinical guideline 127],
Prevention of cardiovascular disease [NICE public health guidance 25] and
Physical activity [NICE public health guidance 44]). [new 2014]

14.5.3.4 Treat people with psychosis or schizophrenia who have diabetes and/or
cardiovascular disease in primary care according to the appropriate NICE
guidance (for example, see Lipid modification [NICE clinical guideline 67],
Type 1 diabetes [NICE clinical guideline 15], Type 2 diabetes [NICE clinical
guideline 66], Type 2 diabetes – newer agents [NICE clinical guideline 87]).
[2009]
14.5.3.5 Healthcare professionals in secondary care should ensure, as part of the care
programme approach, that people with psychosis or schizophrenia receive
physical healthcare from primary care as described in recommendations
14.5.3.1–14.5.3.4. [2009]
Relapse and re-referral to secondary care

14.5.3.6 When a person with an established diagnosis of psychosis or schizophrenia
presents with a suspected relapse (for example, with increased psychotic
symptoms or a significant increase in the use of alcohol or other substances),
primary healthcare professionals should refer to the crisis section of the care
plan. Consider referral to the key clinician or care coordinator identified in
the crisis plan. [2009]
14.5.3.7 For a person with psychosis or schizophrenia being cared for in primary
care, consider referral to secondary care again if there is:
• poor response to treatment
• non-adherence to medication
• intolerable side effects from medication
• comorbid substance misuse
• risk to self or others. [2009]
14.5.3.8 When re-referring people with psychosis or schizophrenia to mental health
services, take account of service user and carer requests, especially for:
• review of the side effects of existing treatments
• psychological treatments or other interventions. [2009]
Transfer
14.5.3.9 When a person with psychosis or schizophrenia is planning to move to the
catchment area of a different NHS trust, a meeting should be arranged
between the services involved and the service user to agree a transition plan
before transfer. The person’s current care plan should be sent to the new
secondary care and primary care providers. [2009]
14.5.4 Psychological interventions

14.5.4.1 Offer CBT to assist in promoting recovery in people with persisting positive
and negative symptoms and for people in remission. Deliver CBT as
described in recommendation 14.3.7.1. [2009]

14.5.4.2 Offer family intervention to families of people with psychosis or
schizophrenia who live with or are in close contact with the service user.
Deliver family intervention as described in recommendation 14.3.7.2. [2009]
14.5.4.3 Family intervention may be particularly useful for families of people with
psychosis or schizophrenia who have:
• recently relapsed or are at risk of relapse
• persisting symptoms. [2009]
14.5.4.4 Consider offering arts therapies to assist in promoting recovery, particularly
in people with negative symptoms. [2009]
14.5.5 Pharmacological interventions

14.5.5.1 The choice of drug should be influenced by the same criteria recommended
for starting treatment (see sections 14.3.5.and 14.3.6). [2009]
14.5.5.2 Do not use targeted, intermittent dosage maintenance strategies 60 routinely.
However, consider them for people with psychosis or schizophrenia who are
unwilling to accept a continuous maintenance regimen or if there is another
contraindication to maintenance therapy, such as side-effect sensitivity.
[2009]
14.5.5.3 Consider offering depot /long-acting injectable antipsychotic medication to
people with psychosis or schizophrenia:
• who would prefer such treatment after an acute episode
• where avoiding covert non-adherence (either intentional or
unintentional) to antipsychotic medication is a clinical priority
within the treatment plan. [2009]
14.5.6 Using depot/long-acting injectable antipsychotic medication

14.5.6.1 When initiating depot/long-acting injectable antipsychotic medication:
• take into account the service user’s preferences and attitudes
towards the mode of administration (regular intramuscular
injections) and organisational procedures (for example, home visits
and location of clinics)
• take into account the same criteria recommended for the use of oral
antipsychotic medication (see sections 14.3.5 and 14.3.6),
particularly in relation to the risks and benefits of the drug regimen
• initially use a small test dose as set out in the BNF or SPC. [2009]
60 Defined as the use of antipsychotic medication only during periods of incipient relapse or symptom
exacerbation rather than continuously.

14.5.7 Interventions for people whose illness has not responded
adequately to treatment
14.5.7.1 For people with schizophrenia whose illness has not responded adequately
to pharmacological or psychological treatment:
• Review the diagnosis.
• Establish that there has been adherence to antipsychotic
medication, prescribed at an adequate dose and for the correct
duration.
• Review engagement with and use of psychological treatments and
ensure that these have been offered according to this guideline. If
family intervention has been undertaken suggest CBT; if CBT has
been undertaken suggest family intervention for people in close
contact with their families.
• Consider other causes of non-response, such as comorbid
substance misuse (including alcohol), the concurrent use of other
prescribed medication or physical illness. [2009]
14.5.7.2 Offer clozapine to people with schizophrenia whose illness has not
responded adequately to treatment despite the sequential use of adequate
doses of at least 2 different antipsychotic drugs. At least 1 of the drugs
should be a non-clozapine second-generation antipsychotic. [2009]
14.5.7.3 For people with schizophrenia whose illness has not responded adequately
to clozapine at an optimised dose, healthcare professionals should consider
recommendation 14.5.7.1(including measuring therapeutic drug levels)
before adding a second antipsychotic to augment treatment with clozapine.
An adequate trial of such an augmentation may need to be up to 8–10 weeks.
Choose a drug that does not compound the common side effects of
clozapine. [2009]
14.5.8 Employment, education and occupational activities

14.5.8.1 Offer supported employment programmes to people with psychosis or
schizophrenia who wish to find or return to work. Consider other
occupational or educational activities, including pre-vocational training, for
people who are unable to work or unsuccessful in finding employment.
[new 2014]
14.5.8.2 Mental health services should work in partnership with local stakeholders,
including those representing black, Asian and minority ethnic groups, to
enable people with mental health problems, including psychosis or
schizophrenia, to stay in work or education and to access new employment
(including self-employment), volunteering and educational opportunities.
[2009; amended 2014]
14.5.8.3 Routinely record the daytime activities of people with psychosis or
schizophrenia in their care plans, including occupational outcomes. [2009]

14.6 RESEARCH RECOMMENDATIONS
14.6.1 Peer support interventions
What is the clinical and cost effectiveness of peer support interventions in people
with psychosis and schizophrenia?
Why this is important

Service users have supported the development of peer support interventions, which
have recently proliferated in the UK, but current evidence for these interventions in
people with psychotic disorders is not strong and the studies are mainly of very low
quality. Moreover the content of the programmes has varied considerably, some
using structured interventions, others providing more informal support. There is
therefore an urgent need for high-quality evidence in this area.
The programme of research would be in several stages. First, there should be

development work to establish what specifically service users want from peer
support workers, as opposed to what they want from professionals, and what the
conditions are for optimal delivery of the intervention. This development work
should be co-produced by exploring the views of service users, experienced peer
support workers and developers of peer support interventions, and suitable
outcome measures should be identified reflecting the aims of peer support. Second,
the intervention, delivered as far as possible under the optimal conditions, should be
tested in a high-quality trial. Further research should test structured and manualised
formats versus unstructured formats (in which service user and peer decide together
what to cover in the session). Benefits and adverse effects experienced by peer
support workers should also be measured.
14.6.2 People who choose not to take antipsychotic medication

What is the clinical and cost effectiveness of psychological intervention alone,
compared with treatment as usual, in people with psychosis or schizophrenia who
choose not to take antipsychotic medication?

The development of alternative treatment strategies is important for the high
proportion of people with psychosis and schizophrenia who choose not to take
antipsychotic medication, or discontinue it because of adverse effects or lack of
efficacy. There is evidence that psychological interventions (CBT and family
intervention) as an adjunct to antipsychotic medication are effective in the treatment
of psychosis and schizophrenia and are cost saving. However, there is little evidence
for family intervention or CBT alone, without antipsychotic medication.
The programme of research should compare the clinical and cost effectiveness of
psychological intervention alone (CBT and/or family intervention) with treatment as
usual for people with psychosis or schizophrenia who choose not to take

antipsychotic medication, using an adequately powered study with a randomised
controlled design. Key outcomes should include symptoms, relapse rates, quality of
life, treatment acceptability, social functioning and the cost effectiveness of the
interventions.
14.6.3 The physical health benefits of discontinuing antipsychotic

medication
What are the short- and long-term benefits to physical health of guided medication
discontinuation and/or reduction in first episode psychosis and can this be achieved
without major risks?

There is growing concern about the long-term health risks, increased mortality and
cortical grey matter loss linked to cumulative neuroleptic exposure in people with
psychosis. The majority of young adults discontinue their medication in an
unplanned way because of these risks. A Dutch moderately-sized open trial has
reported successful discontinuation of medication in 20% of people without serious
relapse; at 7-year follow-up there was continuous benefit for guided reduction in
terms of side effects, functioning and employment, with no long-term risks. If
replicated, this would mark a significant breakthrough in reducing the long-term
physical health risks associated with antipsychotic treatment and improving
outcomes.
The programme of research should use an adequately powered, multicentre, double-

blind, randomised controlled design to test the physical health benefits, risks and
costs of discontinuing or reducing antipsychotic medication among young adults
with first episode psychosis who have achieved remission. The primary outcomes
should be quality of life and metabolic disorder, including weight gain; secondary
outcomes should include side effects, serious relapse, acceptability and user
preference.
14.6.4 Maintaining the benefits of early intervention in psychosis

services after discharge
How can the benefits of early intervention in psychosis services be maintained once
service users are discharged after 3 years?

Early intervention in psychosis services deliver evidence-based interventions in a
positive, youth-friendly setting, improve outcomes, are cost effective and have high
service user acceptability and engagement. Once people are transferred to primary
care or community mental health services these gains are diminished. The guideline
recommends that trusts consider extending these services. However, the extent to
which gains would be maintained and who would benefit most is not known. The
successful element of early intervention in psychosis services might be incorporated

into mainstream services for psychosis, but how this would function, and its cost
effectiveness, needs to be determined.
The suggested programme of research should use an adequately powered, multi-
centre randomised trial comparing extending early intervention in psychosis
services (for example, for 2 years) versus providing augmented (step-down) care in
community mental health services versus treatment as usual to determine whether
the gains of early intervention can be maintained and which service users would
benefit most under each condition. The primary outcome should be
treatment/service engagement and secondary outcomes should include relapse,
readmission, functioning and user preference.
14.6.5 Interventions for PTSD symptoms in people with psychosis and

schizophrenia
What is the benefit of a CBT-based trauma reprocessing intervention on PTSD
symptoms in people with psychosis and schizophrenia?

PTSD symptoms have been documented in approximately one-third of people with
psychosis and schizophrenia. The absence of PTSD symptoms in this context
predicts better mental health outcomes, lower service use and improved life
satisfaction. Two-thirds of the traumatic intrusions, observed in first episode and
established psychosis, relate to symptoms of psychosis and its treatment (including
detention). One study has demonstrated proof-of-principle in first episode psychosis
for trauma reprocessing, focusing on psychosis-related intrusions. Replication of the
study will fill a major gap in treatment for this population and may have other
benefits on psychotic symptoms and service use.
The suggested programme of research would use an adequately powered, multi-

centre randomised trial to test whether a CBT-based trauma reprocessing
intervention can reduce PTSD symptoms and related distress in people with
psychosis and schizophrenia. The trial should be targeted at those with high levels of
PTSD symptoms, particularly traumatic intrusions, following first episode psychosis.
The follow-up should be up to 2 years and the intervention should include ‘booster’
elements, extra sessions of CBT-based trauma reprocessing interventions, and a
health economic evaluation.

15 REFERENCES
Aagaard J, Freiesleben M, Foldager L. Crisis homes for adult psychiatric patients.
Social Psychiatry and Psychiatric Epidemiology. 2008;43:403-9.
Aberg-Wistedt A, Cressell T, Lidberg Y, Liljenberg B, Osby U. Two-year outcome of

team-based intensive case management for patients with schizophrenia. Psychiatric
Services. 1995;46:1263-66.
Achim AM, Maziade M, Raymond É, Olivier D, Mérette C, Roy M. How prevalent

are anxiety disorders in schizophrenia? A meta-analysis and critical review on a
significant association. Schizophrenia Bulletin. 2011;37:811-21.
Acil AA, Dogan S, Dogan O. The effects of physical exercises to mental state and
quality of life in patients with schizophrenia. Journal of Psychiatric and Mental
Health Nursing. 2008;15:808-15.
Adams CE, Fenton MK, Quraishi S, David AS. Systematic meta-review of depot
antipsychotic drugs for people with schizophrenia. The British Journal of Psychiatry.
2001;179:290-9.
Addington J, Epstein I, Liu L, French P, Boydell KM, Zipursky RB. A randomized

controlled trial of cognitive behavioral therapy for individuals at clinical high risk of
psychosis. Schizophrenia Research. 2011;125:54-61.
Addington J, Saeedi H, Addington D. Weight gain in first-episode psychosis over

three years. Schizophrenia Research. 2006;86:335-36.
Advisory Conciliation and Arbitration Service. ACAS- Promoting employment

relations and HR excellence. 2009.[cited 29 May 12]:Available from:
http://www.acas.org.uk/index.aspx?articleid=1361.
Agid O, Mamo D, Ginovart N, Vitcu I, Wilson AA, Zipursky RB, et al. Striatal vs
extrastriatal dopamine D2 receptors in antipsychotic response-a double-blind PET
study in schizophrenia. Neuropsychopharmacology. 2007;32:1209-15.
AGREE Collaboration. Development and validation of an international appraisal

instrument for assessing the quality of clinical practice guidelines: the AGREE
project. Quality and Safety in Health Care. 2003;12:18-23.
Akbarpour F, Rezaei O, Khodaie-Ardakani MR, Sheikhvatan M, Goodarzi H,

Dolatshahi B. A double-blind placebo-controlled trial of bupropion for smoking

abstinence and cognition improvement in schizophrenia. Minerva Psichiatrica.
2010;51:263-69.
Alanen YO. Schizophrenia: Its Origins and Need-Adapted Treatment. London:

Karnac Books; 1997.
Alberti KG, Zimmet P, Shaw J. The metabolic syndrome--a new worldwide

definition. The Lancet. 2005;366:1059-62.
Alexeyeva I, Mauskopf J, Earnshaw SR, Stauffer VL, Gibson JP, Ascher-Svanum H, et

al. Comparing olanzapine and ziprasidone in the treatment of schizophrenia: a case
study in modeling. Journal of Drug Assessment. 2001;4:275-88.
Almerie MQ, Alkhateeb H, Essali A, Matar HE, Rezk E. Cessation of medication for
people with schizophrenia already stable on chlorpromazine. Cochrane Database of
Systematic Reviews. 2007 Art. No. CD006329.
Almond S, Knapp M, Francois C, Toumi M, Brugha T. Relapse in schizophrenia:

costs, clinical outcomes and quality of life. The British Journal of Psychiatry.
2004;184:346-51.
Almond S, O’Donnell O. Cost analysis of the treatment of schizophrenia in the UK.

Pharmacoeconomics. 2000;17:383-89.
Alonso J, Angermeyer MC, Bernert S, Bruffaerts R, Brugha TS, Bryson H, et al.

Prevalence of mental disorders in Europe: results from the European Study of the
Epidemiology of Mental Disorders (ESEMeD) project. Acta Psychiatrica
Scandinavica Supplementum. 2004:21-7.
Alvarez-Jimenez M, Gonzalez-Blanch C, Crespo-Facorro B, Hetrick S, Rodriguez-

Sanchez JM, Perez-Iglesias R, et al. Antipsychotic-induced weight gain in chronic
and first-episode psychotic disorders: a systematic critical reappraisal. CNS Drugs.
2008;22:547-62.
Alvarez-Jiménez M, González-Blanch C, Vázquez-Barquero JL, Pérez-Iglesias R,

Martínez-García O, Pérez-Pardal T, et al. Attenuation of antipsychotic-induced
weight gain with early behavioral intervention in drug-naive first-episode psychosis
patients: a randomized controlled trial. Journal of Clinical Psychiatry. 2006;67:1253-
60.
American Diabetes Association, American Psychiatric Association, American

Association of Clinical E, North American Association for the Study of Obesity.
Consensus development conference on antipsychotic drugs and obesity and
diabetes. Journal of Clinical Psychiatry. 2004;65:267-72.

Amminger GP, Schafer MR, Papageorgiou K, Klier CM, Cotton SM, Harrigan SM, et
al. Long-chain omega-3 fatty acids for indicated prevention of psychotic disorders: a
randomized, placebo-controlled trial. Archives of General Psychiatry. 2010;67:146-54.
Andrew A, Knapp M, McCrone PR, Parsonage M, Trachtenberg M. Effective

Interventions in Schizophrenia: The Economic Case. London: Personal Social
Services Research Unit, London School of Economics and Political Science; 2012.
Anzai N, Yoneda S, Kumagai N, Nakamura Y, Ikebuchi E, Liberman RP. Training

persons with schizophrenia in illness self-management: a randomized controlled
trial in Japan. Psychiatric Services. 2002;53:545-47.
Armstrong NP, Steffen JJ. The recovery promotion fidelity scale: assessing the
organizational promotion of recovery. Community Mental Health Journal.
2009;45:163-70.
Arseneault L, Cannon M, Witton J, Murray RM. Causal association between cannabis

and psychosis: examination of the evidence. The British Journal of Psychiatry.
2004;184:110-7.
Askey R, Holmshaw J, Gamble C, Gray R. What do carers of people with psychosis

need from mental health services? Exploring the views of carers, service users and
professionals. Journal of Family Therapy. 2009;31:310-31.
Assion HJ, Reinbold H, Lemanski S, Basilowski M, Juckel G. Amisulpride

augmentation in patients with schizophrenia partially responsive or unresponsive to
clozapine. A randomized, double-blind, placebo-controlled trial.
Pharmacopsychiatry. 2008;41:24-8.
Attux C, Martini LC, Elkis H, Tamai S, Freirias A, Camargo MDGM, et al. A 6-month
randomized controlled trial to test the efficacy of a lifestyle intervention for weight
gain management in schizophrenia. BMC Psychiatry. 2013;13.
Audini B, Marks IM, Lawrence RE, Connolly J, Watts V. Home-based versus out-
patient/in-patient care for people with serious mental illness. Phase II of a controlled
study. The British Journal of Psychiatry. 1994;165:204-10.
Awad AG, Voruganti LN. The burden of schizophrenia on caregivers: a review.

Ayllon T, Azrin NH. The measurement and reinforcement of behavior of psychotics.

Journal of the Experimental Analysis of Behavior. 1965;8:357-83.
Bagnall AM, Jones L, Ginnelly L, Lewis R, Glanville J, Gilbody S, et al. A systematic

review of atypical antipsychotic drugs in schizophrenia. Health Technology
Assessment. 2003;7:1-193.

Baker A, Richmond R, Haile M, Lewin T, Carr V, Taylor R, et al. A randomized
controlled trial of a smoking cessation intervention among people with a psychotic
disorder. American Journal of Psychiatry. 2006;163:1934-42.
Banerjee S, Clancy C, Crome I. Co-existing Problems of Mental Disorder and

Substance Misuse ('Dual Diagnosis'): An Information Manual. London: Royal
College of Psychiatrists Research Unit; 2002.
Banham L, Gilbody S. Smoking cessation in severe mental illness: what works?

Addiction. 2010;105:1176-89.
Barbic S, Krupa T, Armstrong I. A randomized controlled trial of the effectiveness of

a modified recovery workbook program: preliminary findings. Psychiatric Services.
2009;60:491-97.
Barlow J, Wright C, Sheasby J, Turner A, Hainsworth J. Self-management

approaches for people with chronic conditions: a review. Patient Education and
Counseling. 2002;48:177-87.
Barnable A, Gaudine A, Bennett L, Meadus R. Having a sibling with schizophrenia: a

phenomenological study. Research and Theory for Nursing Practice. 2006;20:247-64.
Barnes TR, Curson DA. Long-term depot antipsychotics. A risk-benefit assessment.

Drug Safety. 1994;10:464-79.
Barnes TR, Drake MJ, Paton C. Nocturnal enuresis with antipsychotic medication.
The British Journal of Psychiatry. 2012;200:7-9.
Barnes TR, McPhillips MA. Critical analysis and comparison of the side-effect and
safety profiles of the new antipsychotics. The British Journal of Psychiatry. 1999:34-
43.
Barnes TR, Paton C, Cavanagh MR, Hancock E, Taylor DM, UK Prescribing

Observatory for Mental Health. A UK audit of screening for the metabolic side
effects of antipsychotics in community patients. Schizophrenia Bulletin.
2007;33:1397-403.
Barnes TR, Shingleton-Smith A, Paton C. Antipsychotic long-acting injections:

prescribing practice in the UK. The British Journal of Psychiatry. 2009;52:S37-42.
Barnes TRE, Buckley P, Schulz SC. Treatment-resistant schizophrenia. In: Hirsch SR,
Weinberger DR, eds. Schizophrenia Second Edition. Oxford: Blackwell Publishing;
2003.

Barrowclough C, Haddock G, Tarrier N, Lewis SW, Moring J, O'Brien R, et al.
Randomized controlled trial of motivational interviewing, cognitive behavior
therapy, and family intervention for patients with comorbid schizophrenia and
substance use disorders. American Journal of Psychiatry. 2001;158:1706-13.
Basset T, Faulkner A, Repper J, Stamou E. Lived experience leading the way: peer
support in mental health. Report. London: Together for Mental Wellbeing; 2010.
Bauer MS, McBride L, Williford WO, Glick H, Kinosian B, Altshuler L, et al.

Collaborative care for bipolar disorder: part I. Intervention and implementation in a
randomized effectiveness trial. Psychiatric Services. 2006;57:927-36.
Beard JH, Pitt RB, Fisher SH, Goertzel V. Evaluating the effectiveness of a psychiatric
rehabilitation program. American Journal of Orthopsychiatry. 1963;33:701-12.
Beard SM, Maciver F, Clouth J, Ruther E. A decision model to compare health care
costs of olanzapine and risperidone treatment for schizophrenia in Germany. The
European Journal of Health Economics. 2006;7:165-72.
Bebbington P, Kuipers L. The predictive utility of expressed emotion in

schizophrenia: an aggregate analysis. Psychological medicine. 1994;24:707-18.
Bebbington PE, Bhugra D, Brugha T, Singleton N, Farrell M, Jenkins R, et al.

Psychosis, victimisation and childhood disadvantage: evidence from the second
British national survey of psychiatric morbidity. The British Journal of Psychiatry.
2004;185:220-6.
Bechdolf A, Wagner M, Ruhrmann S, Harrigan S, Putzfeld V, Pukrop R, et al.

Preventing progression to first-episode psychosis in early initial prodromal states.
Beck AT. Cognitive Therapy and the Emotional Disorders. New York: International
Universities Press; 1979.
Becker D, Whitley R, Bailey EL, Drake RE. Long-term employment trajectories

among participants with severe mental illness in supported employment. Psychiatric
Services. 2007;58:922-28.
Becker RE. An evaluation of a rehabilitation program for chronically hospitalized

psychiatric patients. Social Psychiatry and Psychiatric Epidemiology. 1967;2:32-38.
Beebe LH. Effect of a motivational group intervention on exercise self-efficacy and

outcome expectations for exercise in schizophrenia spectrum disorders. Journal of
the American Psychiatric Nurses Association. 2010;16:105-13.

Bell DNF, Blanchflower DG. Young people and the Great Recession. Oxford Review
of Economic Policy. 2011;27:241-67.
Bell M, Lysaker P, Bryson G. A behavioral intervention to improve work

performance in schizophrenia: work behavior inventory feedback. Journal of
Vocational Rehabilitation. 2003;18:43-50.
Bell M, Zito W, Greig T. Neurocognitive enhancement therapy and competitive

employment in schizophrenia: Effects on clients with poor community functioning.
American Journal of Psychiatric Rehabilitation. 2008;11:109-22.
Bell MD, Bryson GJ, Greig TC, Fiszdon JM, Wexler BE. Neurocognitive enhancement
therapy with work therapy: productivity outcomes at 6- and 12-month follow-ups.
Journal of Rehabilitation Research and Development. 2005;42:829-38.
Bell MD, Milstein RM, Lysaker PH. Pay as an incentive in work participation by
patients with severe mental-illness. Hospital and Community Psychiatry.
1993;44:684-86.
Bellack AS. Skills training for people with severe mental illness. Psychiatric
Rehabilitation Journal. 2004;27:375-91.
Bennett D, Freeman H, eds. Principles and prospect. In Community Psychiatry.

Edinburgh: Churchill Livingstone; 1991.
Bentall RP. Deconstructing the concept of ‘schizophrenia’. Journal of Mental Health.

1993;2:223-38.
Bentall RP, Jackson HF, Pilgrim D. Abandoning the concept of 'schizophrenia': some
implications of validity arguments for psychological research into psychotic
phenomena. The British Journal of Clinical Psychology / The British Psychological
Society. 1988;27:303-24.
Bentall RP, Morrison AP. More harm than good: the case against using antipsychotic
drugs to prevent severe mental illness. Journal of Mental Health. 2002;11:351-56.
Bergner E, Leiner AS, Carter T, Franz L, Thompson NJ, Compton MT. The period of
untreated psychosis before treatment initiation: a qualitative study of family
members' perspectives. Comprehensive Psychiatry. 2008;49:530-6.
Berlin JA. Does blinding of readers affect the results of meta-analyses? The Lancet.
2001;350:185−86.
Bertolote J, McGorry P. Early intervention and recovery for young people with
psychosis: consensus statement. British Journal of Psychiatry. 2005;187:116-19.

Bhugra D, Harding C, Lippett R. Pathways into care and satisfaction with primary
care for black patients in South London. Journal of Mental Health. 2004;13:171–83.
Bhui K, Bhugra D, McKenzie K. Specialist Services for Minority Ethnic Groups?

Maudsley Discussion Paper no 8. 2000.
Bhui K, McKenzie K, Gill P. Delivering mental health services for a diverse society.
BMJ. 2004;329:363-4.
Bhui K, Sashidharan SP. Should there be separate psychiatric services for ethnic
minority groups? The British Journal of Psychiatry. 2003;182:10-12.
Bhui K, Warfa N, Edonya P, McKenzie K, Bhugra D. Cultural competence in mental

health care: a review of model evaluations. BMC health services research. 2007;7:15.
Bhui KS, McKenzie K. Rates and risk factors by ethnic group for suicides within a
year of contact with mental health services in England and Wales. Psychiatric
Services. 2008;59:414-20.
Biggs D, Hovey N, Tyson PJ, MacDonald S. Employer and employment agency

attitudes towards employing individuals with mental health needs. Journal of
Mental Health. 2010;19:505-16.
Bindman J, Johnson S, Wright S, Szmukler G, Bebbington P, Kuipers E, et al.

Integration between primary and secondary services in the care of the severely
mentally ill: patients' and general practitioners' views. The British Journal of
Psychiatry. 1997;171:169-74.
Bio DS, Gattaz WF. Vocational rehabilitation improves cognition and negative
symptoms in schizophrenia. Schizophrenia Research. 2011;126:265-69.
Birchwood M. Pathways to emotional dysfunction in first-episode psychosis. The

British Journal of Psychiatry. 2003;182:373-5.
Birchwood M, Smith J, Cochrane R. Specific and non-specific effects of educational

intervention for families living with schizophrenia. A comparison of three methods.
Bjorkman T, Hansson L, Sandlund M. Outcome of case management based on the

strengths model compared to standard care. A randomised controlled trial. Social
Psychiatry and Psychiatric Epidemiology. 2002;37:147-52.
Blankertz L, Robinson S. Adding a vocational focus to mental health rehabilitation.

Psychiatric Services. 1996;47:1216-22.

Bleuler E. Dementia Praecox or the Group of Schizophrenias (translated by J. Zinkin,
1950). New York: International Universities Press; 1911.
Bleuler M. The schizophrenic disorders: Long-term patient and family studies. New
Haven, CT: Yale University Press; 1978.
Bloch B, Reshef A, Cohen T, Tafla A, Gathas S, Israel S, et al. Preliminary effects of

bupropion and the promoter region (HTTLPR) serotonin transporter (SLC6A4)
polymorphism on smoking behavior in schizophrenia. Psychiatry research.
2010;175:38-42.
Boardman J, Grove B, Perkins R, Shepherd G. Work and employment for people

with psychiatric disabilities. The British Journal of Psychiatry. 2003;182:467-8.
Bola JR, Mosher LR. At issue: predicting drug-free treatment response in acute
psychosis from the Soteria project. Schizophrenia Bulletin. 2002;28:559-75.
Bond GR, Dietzen LL, Mcgrew JH, Miller LD. Accelerating entry into supported
employment for persons with severe psychiatric disabilities. Rehabilitation
Psychology. 1995;40:75-94.
Bond GR, Dincin J. Accelerating entry into transitional employment in a

psychosocial rehabilitation agency. Rehabilitation Psychology. 1986;31:143-55.
Bond GR, Drake RE, Becker DR. Generalizability of the Individual Placement and
Support (IPS) model of supported employment outside the US. World Psychiatry.
2012;11:32-9.
Bond GR, Miller LD, Krumwied RD, Ward RS. Assertive case management in three
CMHCs: a controlled study. Hospital and Community Psychiatry. 1988;39:411-8.
Bond GR, Salyers MP, Dincin J, Drake R, Becker DR, Fraser VV, et al. A randomized
controlled trial comparing two vocational models for persons with severe mental
illness. Journal of Consulting and Clinical Psychology. 2007;75:968-82.
Bond GR, Witheridge TF, Dincin J, Wasmer D, Webb J, De Graaf-Kaser R. Assertive

community treatment for frequent users of psychiatric hospitals in a large city: a
controlled study. American Journal of Community Psychology. 1990;18:865-91.
Bondolfi G, Dufour H, Patris M, May JP, Billeter U, Eap CB, et al. Risperidone versus
clozapine in treatment-resistant chronic schizophrenia: a randomized double-blind
study. The Risperidone Study Group. American Journal of Psychiatry. 1998;155:499-
504.

Bottlender R, Sato T, Jager M, Wegener U, Wittmann J, Strauss A, et al. The impact of
the duration of untreated psychosis prior to first psychiatric admission on the 15-
year outcome in schizophrenia. Schizophrenia Research. 2003;62:37-44.
Bounthavong M, Okamoto MP. Decision analysis model evaluating the cost-

effectiveness of risperidone, olanzapine and haloperidol in the treatment of
schizophrenia. Journal of Evaluation in Clinical Practice. 2007;13:453-60.
Bouras N, Tufnell G, Brough DI, Watson JP. Model for the integration of community
psychiatry and primary care. The Journal of the Royal College of General
Practitioners. 1986;36:62-6.
Bradstreet S, Pratt R. Developing peer support worker roles: reflecting on

experiences in Scotland. Mental Health and Social Inclusion. 2010;14:36-41.
Brar JS, Ganguli R, Pandina G, Turkoz I, Berry S, Mahmoud R. Effects of behavioral

therapy on weight loss in overweight and obese patients with schizophrenia or
schizoaffective disorder. Journal of Clinical Psychiatry. 2005;66:205-12.
Brazier J. Describing health. In: J. Brazier., J. Ratcliffe., A. Tsuchiya., Salomon. JA,

eds. In Measuring and Valuing Health Benefits for Economic Evaluation. Oxford &
New York: Oxford University Press; 2007a. p. 55-82.
Brazier J. Methods for obtaining health state values: generic preference-based

measures of health and the alternatives. In: J. Brazier., J. Ratcliffe., A. Tsuchiya.,
Salomon. JA, eds. Measuring and Valuing Health Benefits for Economic Evaluation.
Oxford & New York: Oxford University Press.; 2007b. p. 175-256.
Brazier J, Roberts J, Deverill M. The estimation of a preference-based measure of

health from the SF-36. Journal of Health Economics. 2002;21:271-92.
Brenner HD. On the importance of cognitive disorders in treatment and

rehabilitation. In: J. S. Strauss, W. Boker, H. D. Brenner, eds. In Psychosocial
Treatment of Schizophrenia. Toronto: Hans Huber; 1986.
Brenner HD, Dencker SJ, Goldstein MJ, Hubbard JW, Keegan DL, Kruger G, et al.
Defining treatment refractoriness in schizophrenia. Schizophrenia Bulletin.
1990;16:551-61.
Briggs A, Sculpher M, Claxton C. Making decision models probabilistic. In Decision

Modelling for Health Economic Evaluation. Briggs, A., Sculpher, M., Claxton, C.
edn. New York: Oxford University Press; 2006a. p. 77-120.
Briggs AH, Claxton K, Sculpher MJ. Decision Modelling for Health Economic
Evaluation. New York: Oxford University Press; 2006b.

British Medical Association and the Royal Pharmaceutical Society of Great Britain.
British National Formulary (BNF) 56. London: Pharmaceutical Press; 2008; Available
from: http://www.bnf.org/bnf/bnf/56/104945.htm.
British Psychological Society. New Ways of Working for Applied Psychologists in

Health and Social Care – The End of the Beginning. Leicester: British Psychological
Society, Leicester; 2007.
Brown C, Goetz J, Hamera E. Weight loss intervention for people with serious
mental illness: a randomized controlled trial of the RENEW program. Psychiatric
Services. 2011;62:800-2.
Brown GW, Monck E, Carstairs G, Wing J. Influence of family life on the course of
schizophrenic illness. British Journal of Preventive & Social Medicine. 1962;16:55-68.
Brown GW, Rutter M. The measurement of family activities and relationships: a

methodological study. Human Relations. 1966;19:241-63.
Brown S, Birtwistle J, Roe L, Thompson C. The unhealthy lifestyle of people with

schizophrenia. Psychological medicine. 1999;29:697-701.
Brown S, Kim M, Mitchell C, Inskip H. Twenty-five year mortality of a community

cohort with schizophrenia. The British Journal of Psychiatry. 2010;196:116-21.
Buckley LA, Pettit T, Adams CE. Supportive therapy for schizophrenia. The
Cochrane Library. 2007.
Buckley P, Miller A, Olsen J, Garver D, Miller DD, Csernansky J. When symptoms

persist: clozapine augmentation strategies. Schizophrenia Bulletin. 2001;27:615-28.
Buckley PF, Miller BJ, Lehrer DS, Castle DJ. Psychiatric comorbidities and
schizophrenia. Schizophrenia Bulletin. 2009;35:383-402.
Buckley PF, Miller DD, Singer B, Arena J, Stirewalt EM. Clinicians' recognition of the
metabolic adverse effects of antipsychotic medications. Schizophrenia Research.
2005;79:281-8.
Bunt L. Music Therapy: An Art Beyond Words. London: Routledge; 1994.
Burnett R, Mallett R, Bhugra D, Hutchinson G, Der G, Leff J. The first contact of

patients with schizophrenia with psychiatric services: social factors and pathways to
care in a multi-ethnic population. Psychological medicine. 1999;29:475-83.
Burns T, Catty J, Becker T, Drake RE, Fioritti A, Knapp M, et al. The effectiveness of
supported employment for people with severe mental illness: a randomised
controlled trial. The Lancet. 2007a;370:1146-52.

Burns T, Catty J, Dash M, Roberts C, Lockwood L, Marshall M. Use of intensive case
management to reduce time in hospital in people with severe mental illness:
systematic review and meta-regression. BMJ. 2007b;335:336.
Burns T, Cohen A. Item-of-service payments for general practitioner care of severely

mentally ill persons: does the money matter? British Journal of Clinical
Pharmacology. 1998;48:1415-6.
Burns T, Creed F, Fahy T, Thompson S, Tyrer P, White I. Intensive versus standard

case management for severe psychotic illness: a randomised trial. UK 700 Group.
The Lancet. 1999;353:2185-9.
Bush CT, Langford MW, Rosen P, Gott W. Operation outreach: intensive case
management for severely psychiatrically disabled adults. Hospital and Community
Psychiatry. 1990;41:647-49.
Bushe CJ, Taylor M, Haukka J. Mortality in schizophrenia: a measurable clinical

endpoint. Journal of Psychopharmacology. 2010;24:17-25.
Butzlaff RL, Hooley JM. Expressed emotion and psychiatric relapse: a meta-analysis.
Archives of General Psychiatry. 1998;55:547.
Caemmerer J, Correll CU, Maayan L. Acute and maintenance effects of non-

pharmacologic interventions for antipsychotic associated weight gain and metabolic
abnormalities: a meta-analytic comparison of randomized controlled trials.
Schizophrenia Research. 2012;140:159-68.
Caldwell DM, Ades AE, Higgins JP. Simultaneous comparison of multiple

treatments: combining direct and indirect evidence. BMJ. 2005;331:897-900.
Campion J, Checinski K, Nurse J. Review of smoking cessation treatments for people

with mental illness. Report No.: 1355-5146 2008.
Campion J, Hewitt J, Shiers D, Taylor D. Pharmacy guidance on smoking and mental

health. Forum for Mental Health in Primary Care Report. London: Royal College of
Psychiatrists and the Royal College of General Practitioners; 2010.
Cannon TD, Cadenhead K, Cornblatt B, Woods SW, Addington J, Walker E, et al.

Prediction of psychosis in youth at high clinical risk: a multisite longitudinal study
in North America. Archives of General Psychiatry. 2008;65:28.
Cantor-Graae E, Selten JP. Schizophrenia and migration: a meta-analysis and review.

American Journal of Psychiatry. 2005;162:12-24.

Care Quality Commission. The state of health care and adult social care in England:
an overview of key themes in care 2011/12. Report. London: The Stationary Office;
2012.
Care Services Improvement Partnership. Report on early detection and intervention

for young people at risk of developing psychosis. London: Care Services
Improvement Partnership; 2005.
Carpenter WT. Anticipating DSM-V: should psychosis risk become a diagnostic

class? Schizophrenia Bulletin. 2009;35:841-43.
Carpenter WT, Buchanan RW. Lessons to take home from CATIE. Psychiatric
Services. 2008;59:523-5.
Carr VJ, Neil AL, Halpin SA, Holmes S, Lewin TJ. Costs of schizophrenia and other
psychoses in urban Australia: findings from the low prevalence (psychotic) disorders
study. Australian and New Zealand Journal of Psychiatry. 2003;37:31-40.
Carrà G, Montomoli C, Clerici M, Cazzullo CL. Family interventions for

schizophrenia in Italy: randomized controlled trial. European Archives of Psychiatry
and Clinical Neuroscience. 2007;257:23-30.
Chakos M, Lieberman J, Hoffman E, Bradford D, Sheitman B. Effectiveness of

second-generation antipsychotics in patients with treatment-resistant schizophrenia:
a review and meta-analysis of randomized trials. American Journal of Psychiatry.
2001;158:518-26.
Chan J, Sweeting M. Review: Combination therapy with non-clozapine atypical

antipsychotic medication: a review of current evidence. Journal of
Psychopharmacology. 2007;21:657-64.
Chan SH, Lee SW, Chan IW. TRIP: a psycho-educational programme in Hong Kong
for people with schizophrenia. Occupational Therapy International. 2007;14:86-98.
Chandler D, Meisel J, Hu TW, McGowen M, Madison K. Client outcomes in a three-

year controlled study of an integrated service agency model. Psychiatric Services.
1996;47:1337-43.
Chao PJ. A group randomized trial to examine the feasibility and effects of
pedometer use and self-monitoring of daily walking in people with severe and
persistent mental illnesses. Dissertation Abstracts International. 2010;72.
Chen EY, Tang JY, Hui CL, Chiu CP, Lam MM, Law CW, et al. Three-year outcome
of phase-specific early intervention for first-episode psychosis: a cohort study in
Hong Kong. Early Intervention in Psychiatry. 2011;5:315-23.

Chen HK, Lan TH, Wu BJ. A double-blind randomized clinical trial of different
doses of transdermal nicotine patch for smoking reduction and cessation in long-
term hospitalized schizophrenic patients. European Archives of Psychiatry and
Clinical Neuroscience. 2013;263:75-82.
Cheng LY, Chan S. Psychoeducation program for Chinese family carers of members
with schizophrenia. Western Journal of Nursing Research. 2005;27:583-99.
Chien WT, Chan SWC. One-year follow-up of a multiple-family-group intervention

for Chinese families of patients with schizophrenia. Psychiatric Services.
2004b;55:1276-84.
Chien WT, Norman I, Thompson DR. A randomized controlled trial of a mutual

support group for family caregivers of patients with schizophrenia. International
Journal of Nursing Studies. 2004a;41:637-49.
Chien WT, Thompson DR, Norman I. Evaluation of a peer-led mutual support group
for Chinese families of people with schizophrenia. American Journal of Community
Psychology. 2008;42:122-34.
Chien WT, Wong KF. A family psychoeducation group program for Chinese people
with schizophrenia in Hong Kong. Psychiatric Services. 2007;58:1003-06.
Chinman M, Oberman RS, Young AS. A cluster randomized trial of adding peer
specialists to intensive case management teams in the veterans health
administration. The Journal of Behavioral Health Services & Research. 2013:1-13.
Chiu MYL, Wei GFW, Lee S. Personal tragedy or system failure: a qualitative
analysis of narratives of caregivers of people with severe mental illness in Hong
Kong and Taiwan. The International Journal of Social Psychiatry. 2006;52:413-23.
Chong SA, Mythily S, Verma S. Reducing the duration of untreated psychosis and
changing help-seeking behaviour in Singapore. Social Psychiatry and Psychiatric
Epidemiology. 2005;40:619-21.
Chong SA, Remington G. Clozapine augmentation: safety and efficacy.

Schizophrenia Bulletin. 2000;26:421-40.
Chong SA, Tan CH, Lee HS. Hoarding and clozapine-risperidone combination.
Canadian Journal of Psychiatry. 1996;41:315-16.
Chou KR, Liu SY, Chu H. The effects of support groups on caregivers of patients
with schizophrenia. International Journal of Nursing Studies. 2002;39:713-22.

Chouinard G, Albright PS. Economic and health state utility determinations for
schizophrenic patients treated with risperidone or haloperidol. Journal of Clinical
Chue PS, Heeg B, Buskens E, van Hout BA. Modelling the impact of compliance on
the costs and effects of long-acting risperidone in Canada. Pharmacoeconomics.
2005;23 Suppl. 1:62-74.
Ciompi L, Dauwalder H-P, Maier C, Aebi E, Trütsch K, Kupper Z, et al. The pilot
project 'Soteria Berne': clinical experiences and results. Alternatives to the Hospital
for Acute Psychiatric Treatment. 1995;32:133.
Clarke GN, Herinckx HA, Kinney RF, Paulson RI, Cutler DL, Lewis K, et al.
Psychiatric hospitalizations, arrests, emergency room visits, and homelessness of
clients with serious and persistent mental illness: findings from a randomized trial of
two ACT programs vs. usual care. Mental health services research. 2000;2:155-64.
Clarke P, Gray A, Holman R. Estimating utility values for health states of type 2
diabetic patients using the EQ-5D (UKPDS 62). Medical Decision Making.
2002;22:340-49.
Clarke PM, Gray AM, Briggs A, et al. (On behalf of the UK Prospective Diabetes
Study [UKPDS]). Cost-utility analyses of intensive blood glucose and tight blood
pressure control in type 2 diabetes (UKPDS 72). Diabetologia. 2005;48:868–77.
Clay S, Schell S, Corrigan P, Ralph R. On our own, together. Peer programs for
people with mental illness. Nashville, TN: Vanderbilt University Press; 2005.
Cocchi A, Mapelli V, Meneghelli A, Preti A. Cost-effectiveness of treating first-

episode psychosis: five-year follow-up results from an Italian early intervention
programme. Early Intervention in Psychiatry; Aug 2011203-11.
Coid J. Failure in community care: psychiatry's dilemma. BMJ. 1994;308:805.
Cole E, Levy G, King M, Johnson-Sabine E, Hoar A. Pathways to care for patients

with a first episode of psychosis. A comparison of ethnic groups. The British Journal
of Psychiatry. 1995;167:770-6.
Cole J. The effects of an exercise program on chronically mentally ill outpatients; a

study of symptom reduction, physical fitness, and stress. Dissertation Abstracts
International. 1997:64.
Collins PY, Patel V, Joestl SS, March D, Insel TR, Daar AS, et al. Grand challenges in
global mental health. Nature. 2011;475:27-30.

Conley RR, Buchanan RW. Evaluation of treatment-resistant schizophrenia.
Cook JA, Copeland ME, Jonikas JA, Hamilton MM, Razzano LA, Grey DD, et al.
Results of a randomized controlled trial of mental illness self-management using
wellness recovery action planning. Schizophrenia Bulletin. 2011;38:881-91.
Cook JA, Leff HS, Blyler CR, Gold PB, Goldberg RW, Mueser KT, et al. Results of a
multisite randomized trial of supported employment interventions for individuals
with severe mental illness. Archives of General Psychiatry. 2005;62:505-12.
Cook JA, Steigman P, Pickett S, Diehl S, Fox A, Shipley P, et al. Randomized

controlled trial of peer-led recovery education using Building Recovery of Individual
Dreams and Goals through Education and Support (BRIDGES). Schizophrenia
Research. 2012;136:36-42.
Cookson J, Taylor D, Katona C. Use of Drugs in Psychiatry: The Evidence from

Psychopharmacology. London: Gaskell; 2002.
Copeland ME, Mead S. WRAP and Peer Support for People, Groups and Programs.
Brattleboro: Peach Press; 2004.
Corcoran CM, First MB, Cornblatt B. The psychosis risk syndrome and its proposed
inclusion in the DSM-V: a risk-benefit analysis. Schizophrenia Research. 2010;120:16-
22.
Cozolino LJ, Goldstein MJ, Nuechterlein KH, West KL, Snyder KS. The impact of
education about schizophrenia on relatives varying in expressed emotion.
Schizophrenia Bulletin. 1988:675-87.
Craig T, Doherty I, Jamieson-Craig R, Boocock A, Attafua G. The consumer-

employee as a member of a mental health assertive outreach team. I. Clinical and
social outcomes. Journal of Mental Health. 2004a;13:59-69.
Craig TK, Garety P, Power P, Rahaman N, Colbert S, Fornells-Ambrojo M, et al. The

Lambeth early onset (LEO) team: randomised controlled trial of the effectiveness of
specialised care for early psychosis. BMJ. 2004;329:1067.
Crammer J, Eccleston DA. Survey of the use of depot neuroleptics in a whole region.
Psychiatric Bulletin. 1989;13:517–20.
Crawford MJ, Patterson S. Arts therapies for people with schizophrenia: an

emerging evidence base. Evidence Based Mental Health. 2007;10:69-70.

Creamer M, Burgess P, McFarlane AC. Post-traumatic stress disorder: findings from
the Australian National Survey of Mental Health and Well-being. Psychological
medicine. 2001;31:1237-47.
Creed F, Black D, Anthony P, Osborn M, Thomas P, Tomenson B. Randomised

controlled trial of day patient versus inpatient psychiatric treatment. BMJ.
1990;300:1033-7.
Cresswell J, Lelliott P, eds. Accreditation for Inpatient Mental Health Services

(AIMS): National Report for Working Age Acute Wards - July 2007-July 2009.
London: Royal College of Psychiatrists Centre for Quality Improvement; 2009.
Crown S. Supportive psychotherapy: a contradiction in terms? British Journal of

Psychiatry. 1988;152:266-9.
Crowther R, Marshall M, Bond G, Huxley P. Vocational rehabilitation for people

with severe mental illness. Cochrane Database of Systematic Reviews. 2001:Art.
No.:CD003080.
Cummins C, Stevens A, Kisely S, Executive N. The Use of Olanzapine as a First and

Second Choice Treatment in Schizophrenia: Wessex Institute for Health Research
and Development, Development & Evaluation Committee; 1998.
Curtis J, Newall HD, Samaras K. The heart of the matter: cardiometabolic care in
youth with psychosis. Early Intervention in Psychiatry. 2012;6:347-53.
Curtis JL, Millman EJ, Struening E, D'Ercole A. Effect of case management on

rehospitalization and utilization of ambulatory care services. Hospital and
Community Psychiatry. 1992;43:895-9.
Curtis L. Unit Costs of Health and Social Care 2007. Canterbury: Personal Social
Services Research Unit, University of Kent; 2007.
Curtis L. Unit Costs of Health and Social Care 2012. Canterbury: University of Kent;
2012.
Dalack GW, Meador-Woodruff JH. Acute feasibility and safety of a smoking

reduction strategy for smokers with schizophrenia. Nicotine & Tobacco Research.
1999;1:53-7.
Darves-Bornoz J, Lemperiere T, Degiovanni A, Gaillard P. Sexual victimization in

women with schizophrenia and bipolar disorder. Social Psychiatry and Psychiatric
Epidemiology. 1995;30:78-84.
Darves-Bornoz JM, Alonso J, de Girolamo G, de Graaf R, Haro JM, Kovess-Masfety

V, et al. Main traumatic events in Europe: PTSD in the European study of the

epidemiology of mental disorders survey. Journal of Traumatic Stress. 2008;21:455-
62.
Darzi AR. High Quality Care for All: NHS Next Stage Review Final Report. London:
Department of Health; 2008; Available from:
http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPol
icyAndGuidance/DH_085825
Daumit GL, Dickerson FB, Wang N-Y, Dalcin A, Jerome GJ, Anderson CAM, et al. A
behavioral weight-loss intervention in persons with serious mental Illness. New
England Journal of Medicine. 2013;368:1594-602.
David A, Adams CE, Quraishi SN. Depot flupenthixol decanoate for schizophrenia
or other similar psychotic disorders. (Cochrane review). 1999:Art. No.: CD001470.
David AS, Adams C. Depot antipsychotic medication in the treatment of patients

with schizophrenia: (1) meta-review; (2) patient and nurse attitudes. Health
Technology Assessment. 2001;5:1-61.
Davidson L. Supported socialization for people with psychiatric disabilities: lessons

from a randomized controlled trial. Journal of Community Psychology. 2004;32:453.
Davidson L, Chinman M, Kloos B, Weingarten R, Stayner D, Tebes JK. Peer support

among individuals with severe mental illness: A review of the evidence. Clinical
Psychology: Science and Practice. 1999;6:165-87.
Davies A, Langley PC, Keks NA, Catts SV, Lambert T, Schweitzer I. Risperidone
versus haloperidol: II. Cost-effectiveness. Clinical Therapeutics. 1998;20:196-213.
Davies G, Welham J, Chant D, Torrey EF, McGrath J. A systematic review and meta-
analysis of Northern Hemisphere season of birth studies in schizophrenia.
Davies L, Lewis S. Antipsychotic medication for people with first episode

schizophrenia: an exploratory economic analysis of alternative treatment algorithms.
University of York: Centre for Health Economics; 2000.
Davies LM, Barnes TR, Jones PB, Lewis S, Gaughran F, Hayhurst K, et al. A
randomised controlled trial of the cost-utility of second-generation antipsychotics in
people with psychosis and eligible for clozapine. Value in Health. 2008;11:549-62.
Davies LM, Drummond MF. Economics and schizophrenia: the real cost. The British
Journal of Psychiatry. 1994:18-21.
Davies LM, Lewis S, Jones PB, Barnes TR, Gaughran F, Hayhurst K, et al. Cost-
effectiveness of first- v. second-generation antipsychotic drugs: results from a

randomised controlled trial in schizophrenia responding poorly to previous therapy.
British Journal Psychiatry. 2007;191:14-22.
Davis JM, Garver DL. Neuroleptics: Clinical Use in Psychiatry. In: Iversen. L,
Iversen. S, Snyder. S, eds. Handbook of Psychopharmacology. New York: Plenum
Press; 1978 p. 129-64.
Davis JM, Kane JM, Marder SR, Brauzer B, Gierl B, Schooler N, et al. Dose response
of prophylactic antipsychotics. Journal of Clinical Psychiatry. 1993;54 Suppl:24-30.
De Graeve D, Smet A, Mehnert A, Caleo S, Miadi-Fargier H, Mosqueda GJ, et al.

Long-acting risperidone compared with oral olanzapine and haloperidol depot in
schizophrenia: a Belgian cost-effectiveness analysis. Pharmacoeconomics. 2005;23
Suppl 1:35-47.
De Hert M, Dekker JM, Wood D, Kahl KG, Holt RI, Moller HJ. Cardiovascular
disease and diabetes in people with severe mental illness: position statement from
the European Psychiatric Association (EPA), supported by the European Association
for the Study of Diabetes (EASD) and the European Society of Cardiology (ESC).
European Psychiatry. 2009a;24:412-24.
De Hert M, Schreurs V, Vancampfort D, Van Winkel R. Metabolic syndrome in

people with schizophrenia: a review. World Psychiatry. 2009b;8:15-22.
de Leon J, Diaz FJ, Josiassen RC, Cooper TB, Simpson GM. Does clozapine decrease
smoking? Progress in Neuro-psychopharmacology and Biological Psychiatry.
2005;29:757-62.
Deegan P. Recovery as a journey of the heart. Psychiatric Rehabilitation Journal.

1996;19:91-97.
Department of Health. National Service Framework for Mental Health services:

Modern Standards and Service Models. London: Department of Health; 1999;
Available from:
http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPol
icyAndGuidance/DH_4009598
Department of Health. The NHS Plan: a Plan for Investment, a Plan for Reform.
London: Department of Health; 2000.
Department of Health. The Mental Health Policy Implementation Guide. London:

Department of Health; 2001.
Department of Health. Delivering Race Equality: a Framework for Action. London:

Department of Health; 2003; Available from:
www.doh.gov.uk/deliveringraceequality/77951- del_race_equality.pdf.

Department of Health. Delivering Race Equality in Mental Health Care: an Action
Plan for Reform Inside and Outside Services and the Government’s Response to the
Independent Inquiry Into the Death Of David Bennett. London: Department of
Health; 2005; Available from:
http://www.dh.gov.uk/assetRoot/04/10/07/75/04100775.pdf.
Department of Health. NHS Reference Costs 2006–07. London: Department of

Health; 2008; Available from: www.hesonline.nhs.uk.
Dettling M, Sachse C, Brockmoller J, Schley J, Muller-Oerlinghausen B, Pickersgill I,

et al. Long-term therapeutic drug monitoring of clozapine and metabolites in
psychiatric in- and outpatients. Psychopharmacology. 2000;152:80-6.
Diaz FJ, de Leon J, Josiassen RC, Cooper TB, Simpson GM. Plasma clozapine
concentration coefficients of variation in a long-term study. Schizophrenia Research.
2005;72:131-5.
Dick P, Cameron L, Cohen D, Barlow M, Ince A. Day and full time psychiatric
treatment: a controlled comparison. The British Journal of Psychiatry. 1985;147:246-9.
Dick S. Homophobic hate crime: The Gay British Crime Survey. London: Stonewall;
2008.
Dieterich M, Irving CB, Park B, Marshall M. Intensive case management for severe
mental illness. Cochrane Database of Systematic Reviews. 2010;Art. No.:CD007906.
Dincin J, Witheridge TF. Psychiatric rehabilitation as a deterrent to recidivism.

Hospital and Community Psychiatry. 1982;33:645-50.
Dinesh M, David A, Quraishi SN. Depot pipotiazine palmitate and undecylenate for
schizophrenia. Cochrane Database of Systematic Reviews. 2004;Art. No.:CD001720.
Dixon L, Hoch JS, Clark R, Bebout R, Drake R, McHugo G, et al. Cost-effectiveness of

two vocational rehabilitation programs for persons with severe mental illness.
Dixon LB, Lehman AF, Levine J. Conventional antipsychotic medications for

Dowell DA, Ciarlo JA. Overview of the community mental health centers program
from an evaluation perspective. Community Mental Health Journal. 1983;19:95-125.
Drake RE, Becker DR, Anthony WA. A research induction group for clients entering
a mental health research project. Hospital and Community Psychiatry. 1994;45:487-9.

Drake RE, McHugo G. Assertive community treatment for patients with co-
occurring severe mental illness and substance use disorder: a clinical trial. American
Journal of Orthopsychiatry. 1998;68.
Drake RE, McHugo GJ, Bebout RR, Becker DR, Harris M, Bond GR, et al. A
randomized clinical trial of supported employment for inner-city patients with
severe mental disorders. Archives of General Psychiatry. 1999;56:627-33.
Duffy SA, Kilbourne AM, Austin KL, Dalack GW, Woltmann EM, Waxmonsky J, et
al. Risk of smoking and receipt of cessation services among veterans with mental
disorders. Psychiatric Services. 2012;63:126-33.
Dumouchel W, Fram D, Yang XG, Mahmoud RA, Grogg AL, Engelhart L, et al.
Antipsychotics, glycemic disorders, and life-threatening diabetic events: a bayesian
data-mining analysis of the FDA adverse event reporting system (1968-2004). Annals
of Clinical Psychiatry. 2008;20:21-31.
Duraiswamy G, Thirthalli J, Nagendra HR, Gangadhar BN. Yoga therapy as an add-

on treatment in the management of patients with schizophrenia-a randomized
controlled trial. Acta Psychiatrica Scandinavica. 2007;116:226-32.
Durham RC, Guthrie M, Morton RV, Reid DA, Treliving LR, Fowler D, et al.
Tayside-Fife clinical trial of cognitive-behavioural therapy for medication-resistant
psychotic symptoms. Results to 3-month follow-up. The British Journal of
Psychiatry. 2003;182:303-11.
Durson SM, Deakin JF. Augmenting antipsychotic treatment with lamotrigine in

patients with treatment-resistant schizophrenia: a naturalistic case-series outcome
study. Journal of Psychopharmacology. 2001;15:297-301.
Duvivier BM, Schaper NC, Bremers MA, van Crombrugge G, Menheere PP, Kars M,
et al. Minimal intensity physical activity (standing and walking) of longer duration
improves insulin action and plasma lipids more than shorter periods of moderate to
vigorous exercise (cycling) in sedentary subjects when energy expenditure is
comparable. PLoS One. 2013;8:e55542.
Eccles M, Freemantle N, Mason J. North of England evidence based guideline

development project: methods of developing guidelines for efficient drug use in
primary care. BMJ. 1998;316:1232−35.
Eckman TA, Wirshing WC, Marder SR, Liberman RP, Johnston CK, Zimmermann K,
et al. Technique for training schizophrenic patients in illness self-management: a
controlled trial. American Journal of Psychiatry. 1992;149:1549-55.

Edgell ET, Andersen SW, Johnstone BM, Dulisse B, Revicki D, Breier A. Olanzapine
versus risperidone. A prospective comparison of clinical and economic outcomes in
schizophrenia. Pharmacoeconomics. 2000;18:567-79.
Edmundson E, Bedell J, Archer R, Gordon R. Intergrating skill building and peer

support in mental health treatment. In: Jeger A, Slotnick R., ed. Community Mental
Health and Behavioural Ecology. NY: Springer US; 1982. p. 127-39.
Edwards NC, Locklear JC, Rupnow MF, Diamond RJ. Cost effectiveness of long-
acting risperidone injection versus alternative antipsychotic agents in patients with
schizophrenia in the USA. Pharmacoeconomics. 2005;23 Suppl 1:75-89.
Essock SM, Kontos N. Implementing assertive community treatment teams.

Essock SM, Mueser KT, Drake RE, Covell NH, McHugo GJ, Frisman LK, et al.
Comparison of ACT and standard case management for delivering integrated
treatment for co-occurring disorders. Psychiatric Services. 2006;57:185-96.
Evans J, Repper J. Employment, social inclusion and mental health. Journal of

Psychiatric and Mental Health Nursing. 2000;7:15-24.
Evans S, Newton R, Higgins S. Nutritional intervention to prevent weight gain in

patients commenced on olanzapine: a randomized controlled trial. Australian and
New Zealand Journal of Psychiatry. 2005;39:479-86.
Evins AE, Cather C, Culhane MA, Birnbaum A, Horowitz J, Hsieh E, et al. A 12-week
double-blind, placebo-controlled study of bupropion sr added to high-dose dual
nicotine replacement therapy for smoking cessation or reduction in schizophrenia.
Journal of Clinical Psychopharmacology. 2007;27:380-6.
Evins AE, Cather C, Deckersbach T, Freudenreich O, Culhane MA, Olm-Shipman

CM, et al. A double-blind placebo-controlled trial of bupropion sustained-release for
smoking cessation in schizophrenia. Journal of Clinical Psychopharmacology.
2005;25:218-25.
Evins AE, Mays VK, Rigotti NA, Tisdale T, Cather C, Goff DC. A pilot trial of
bupropion added to cognitive behavioral therapy for smoking cessation in
schizophrenia. Nicotine & Tobacco Research. 2001;3:397-403.
Expert Group. Schizophrenia and diabetes 2003: an expert consensus meeting.

Introduction. The British Journal of Psychiatry. 2004;47:S53-4.
Faber G, Smid HGOM, VanGool AR, Wiersma D, VanDen Bosch RJ. The effects of
guided discontinuation of antipsychotics on neurorecognition in first onset
psychosis. European Psychiatry. 2012;27:275-80.

Fadden G, Heelis R. The Meriden Family Programme: lessons learned over 10 years.
Journal of Mental Health. 2011;20:79-88.
Farde L, Nordstrom AL, Wiesel FA, Pauli S, Halldin C, Sedvall G. Positron emission
tomographic analysis of central D1 and D2 dopamine receptor occupancy in patients
treated with classical neuroleptics and clozapine. Relation to extrapyramidal side
effects. Archives of General Psychiatry. 1992;49:538-44.
Färdig R, Lewander T, Melin L, Folke F, Fredriksson A. A randomized controlled

trial of the illness management and recovery program for persons with
schizophrenia. Psychiatric Services. 2011;62:606-12.
Fatemi SH, Folsom TD. The neurodevelopmental hypothesis of schizophrenia,

revisited. Schizophrenia Bulletin. 2009;35:528-48.
Fatemi SH, Stary JM, Hatsukami DK, Murphy SE. A double-blind placebo-controlled
cross over trial of bupropion in smoking reduction in schizophrenia. Schizophrenia
Research. 2005;76:353-56.
Faulkner A, Basset T. A helping hand: taking peer support into the 21st century.
Mental Health and Social Inclusion. 2012;16:41-47.
Faulkner G, Cohn T, Remington G. Validation of a physical activity assessment tool

for individuals with schizophrenia. Schizophrenia Research. 2006;82:225-31.
Fedorowicz VJ, Fombonne E. Metabolic side effects of atypical antipsychotics in

children: a literature review. Journal of Psychopharmacology. 2005;19:533-50.
Fenton FR, Tessier L, Struening EL. A comparative trial of home and hospital
psychiatric care. One-year follow-up. Archives of General Psychiatry. 1979;36:1073-
79.
Fenton WS, McGlashan TH. Schizophrenia: individual psychotherapy. In

Comprehensive Textbook of Psychiatry. H. Kaplan & B. Sadock edn. Baltimore,
Maryland: Williams & Wilkins.; 1995.
Fenton WS, McGlashan TH. We can talk: Individual psychotherapy for

schizophrenia. American Journal of Psychiatry. 1997;154:1493-95.
Fenwick E, Claxton K, Sculpher M. Representing uncertainty: the role of cost-

effectiveness acceptability curves. Health Economics. 2001;10:779-87.
Fidler JA, Shahab L, West O, MJarvis MJ, McEwen A, Stapleton A, et al. 'The
smoking toolkit study': a national study of smoking cessation in England. BMC
Public Health. 2011;11:479.

Flanagan RJ. Therapeutic monitoring of antipsychotic drugs. CPD Clinical
Biochemistry. 2006:3–18.
Fleischhacker WW, Hummer M. Drug treatment of schizophrenia in the 1990s.

Achievements and future possibilities in optimising outcomes. Drugs. 1997;53:915-
29.
Fleischhacker WW, Siu CO, Boden R, Pappadopulos E, Karayal ON, Kahn RS.
Metabolic risk factors in first-episode schizophrenia: baseline prevalence and course
analysed from the European first-episode schizophrenia trial. The International
Journal of Neuropsychopharmacology. 2012;1:1-9.
Foley DL, Morley KI. Systematic review of early cardiometabolic outcomes of the
first treated episode of psychosis. Archives of General Psychiatry. 2011;68:609-16.
Ford R, Beadsmoore A, Ryan P, Repper J, Craig T, Muijen M. Providing the safety

net: case management for people with a serious mental illness. Journal of Mental
Health. 1995;4:91-97.
Foster K, Meltzer H, Gill B, Hinds K. OPCS Surveys of Psychiatric Morbidity in

Great Britain. London: Office of Population Censuses and Surveys; 1996.
Freeman D, Garety PA. Connecting neurosis and psychosis: the direct influence of
emotion on delusions and hallucinations. Behaviour Research and Therapy.
2003;41:923-47.
Freud S. New introductory lectures on psycho-analysis. In: Strachey J, ed. The

Standard Edition of the Complete Psychological Works of Sigmund Freud. London:
Hogarth Press; 1964.
Frey W, Drake RE, Goldman HH, Salkever D, Miller A, Bond GR. The mental health
treatment study: final report to social security administration. Rockville, MD:
Westat; 2011.
Friedlander AH, Marder SR. The psychopathology, medical management and dental
implications of schizophrenia. Journal of the American Dental Association.
2002;133:603-10.
Frith CD. The Cognitive Neuropsychology of Schizophrenia. Hillsdale, NJ: Lawrence

Erlbaum; 1992.
Fromm-Reichmann F. Principles of Intensive Psychotherapy. Chicago, Il: University

of Chicago Press; 1950.

Gabbard GO. Psychodynamic Psychiatry in Clinical Practice: The DSM-IV Edition.
Washington DC: American Psychiatric Press; 1994.
Gaertner I, Gaertner HJ, Vonthein RI, Dietz K. Therapeutic drug monitoring of

clozapine in relapse prevention: a five-year prospective study. Journal of Clinical
Gallagher SM, Penn PE, Schindler E, Layne W. A comparison of smoking cessation

treatments for persons with schizophrenia and other serious mental illnesses.
Journal of Psychoactive Drugs. 2007;39:487-97.
Galletly CA, Clark CR, MacFarlane AC. Treating cognitive dysfunction in patients
with schizophrenia. Journal of Psychiatry and Neuroscience. 2000;25:117-24.
Ganguly R, Miller LS, Martin BC. Future employability, a new approach to cost-
effectiveness analysis of antipsychotic therapy. Schizophrenia Research. 2003;63:111-
9.
Garety PA, Bebbington P, Fowler D, Freeman D, Kuipers E. Implications for

neurobiological research of cognitive models of psychosis: a theoretical paper.
Psychological medicine. 2007;37:1377-91.
Garety PA, Fowler DG, Freeman D, Bebbington P, Dunn G, Kuipers E. Cognitive--

behavioural therapy and family intervention for relapse prevention and symptom
reduction in psychosis: randomised controlled trial. The British Journal of
Psychiatry. 2008;192:412-23.
Garety PA, Hemsley DR. Delusions: Investigations into the Psychology of Delusional
Reasoning. Hove, UK: Psychology Press; 1994.
Garety PA, Kuipers E, Fowler D, Freeman D, Bebbington PE. A cognitive model of

the positive symptoms of psychosis. Psychological medicine. 2001;31:189-95.
Gater R, Goldberg D, Jackson G, Jennett N, Lowson K, Ratcliffe J, et al. The care of

patients with chronic schizophrenia: a comparison between two services.
Geddes J, Freemantle N, Harrison P, Bebbington P. Atypical antipsychotics in the

treatment of schizophrenia: systematic overview and meta-regression analysis. BMJ.
2000;321:1371-6.
Geitona M, Kousoulakou H, Ollandezos M, Athanasakis K, Papanicolaou S,

Kyriopoulos I. Costs and effects of paliperidone extended release compared with
alternative oral antipsychotic agents in patients with schizophrenia in Greece: a cost
effectiveness study. Annals of general psychiatry. 2008;7:16.

Gelkopf M, Noam S, Rudinski D, Lerner A, Behrbalk P, Bleich A, et al.
Nonmedication smoking reduction program for inpatients with chronic
schizophrenia: a randomized control design study. The Journal of Nervous and
Mental Disease. 2012;200:142-46.
Genc Y, Taner E, Candansayar S. Comparison of clozapine-amisulpride and

clozapine-quetiapine combinations for patients with schizophrenia who are partially
responsive to clozapine: a single-blind randomized study. Advances in Therapy.
2007;24:1-13.
George TP, Vessicchio JC, Sacco KA, Weinberger AH, Dudas MM, Allen TM, et al. A
placebo-controlled trial of bupropion combined with nicotine patch for smoking
cessation in schizophrenia. Biological Psychiatry. 2008;63:1092-6.
George TP, Vessicchio JC, Termine A, Bregartner TA, Feingold A, Rounsaville BJ, et
al. A placebo controlled trial of bupropion for smoking cessation in schizophrenia.
Biological Psychiatry. 2002;52:53-61.
George TP, Ziedonis DM, Feingold A, Pepper WT, Satterburg CA, Winkel J, et al.
Nicotine transdermal patch and atypical antipsychotic medications for smoking
cessation in schizophrenia. American Journal of Psychiatry. 2000;157:1835-42.
Gervey R, Bedell JR. Psychological Assessment And Treatment Of Persons With

Severe Mental Disorders. In: Bendall JR, ed. Supported employment in vocational
rehabilitation. Washington DC: Taylor and Francis; 1994. p. 170-5.
Gilbert PL, Harris MJ, McAdams LA, Jeste DV. Neuroleptic withdrawal in
schizophrenic patients. A review of the literature. Archives of General Psychiatry.
1995;52:173-88.
Gilburt H, Slade M, Rose D, Lloyd-Evans B, Johnson S, Osborn DP. Service users'

experiences of residential alternatives to standard acute wards: qualitative study of
similarities and differences. The British Journal of Psychiatry. 2010;197:s26-s31.
Gillies CL, Lambert PC, Abrams KR, Sutton AJ, Cooper NJ, Hsu RT, et al. Different
strategies for screening and prevention of type 2 diabetes in adults: cost effectiveness
analysis. BMJ. 2008;336.
Gilmore JH. Understanding what causes schizophrenia: a developmental

perspective. American Journal of Psychiatry. 2010;167:8-10.
Gilroy A, McNeilly G. The Changing Shape of Art Therapy. London: Jessica

Kingsley Publishers; 2000.
Gingerich S, Tornvall K. Illness Management and Recovery. Hanover: Hazelden

Publishing; 2005.

Glennie JL. Pharmacoeconomic Evaluation in Schizophrenia: Clozapine in
Treatment-Resistant Schizophrenia and Risperidone in Chronic Schizophrenia.
Ottawa, Canada: Canadian Coordinating Office for Health Technology Assessment;
1997.
Glenny AM, Altman DG, Song F, Sakarovitch C, Deeks JJ, D'Amico R, et al. Indirect
comparisons of competing interventions. Health Technology Assessment. 2005;9:1-
134.
Glover G, Arts G, Babu KS. Crisis resolution/home treatment teams and psychiatric
admission rates in England. The British Journal of Psychiatry. 2006;189:441-5.
Godleski LS, Sernyak MJ. Agranulocytosis after addition of risperidone to clozapine

treatment. American Journal of Psychiatry. 1996;153:735-6.
Goeree R, Farahati F, Burke N, Blackhouse G, O'Reilly D, Pyne J, et al. The economic

burden of schizophrenia in Canada in 2004. Current Medical Research and Opinion.
2005;21:2017-28.
Gold PB, Meisler N, Santos AB, Carnemolla MA, Williams OH, Keleher J.
Randomized trial of supported employment integrated with assertive community
treatment for rural adults with severe mental illness. Schizophrenia Bulletin.
2006;32:378-95.
Goldstein MJ. Psychoeducational family programs in the United States. In:

Moscarelli M, Rupp A, Sartorius N, eds. Handbook of Mental Health Economics and
Health Policy, vol 1: Schizophrenia. New York: John Wiley; 1996.
Goodwin V, Happell B. Conflicting agendas between consumers and carers: The

perspectives of carers and nurses. International Journal of Mental Health Nursing.
2006;15:135-43.
Grawe RW, Falloon IR, Widen JH, Skogvoll E. Two years of continued early
treatment for recent-onset schizophrenia: a randomised controlled study. Acta
Psychiatrica Scandinavica. 2006;114:328-36.
Gray JA, Feldon J, Rawlins JNP, Hemsley DR, Smith AD. The neuropsychology of
schizophrenia. Behavioral and Brain Sciences. 1991;14:1–20.
Gray R, Leese M, Bindman J, Becker T, Burti L, David A, et al. Adherence therapy for
people with schizophrenia. European multicentre randomised controlled trial. The
Green BL, Wehling C, Talsky GJ. Group art-therapy as an adjunct to treatment for
chronic outpatients. Hospital and Community Psychiatry. 1987;38:988-91.

Green MF. Neuropsychological performance in the unaffected twin. Archives of
General Psychiatry. 1992;49:247.
Griffiths RD. Rehabilitation of chronic psychotic patients. An assessment of their

psychological handicap, an evaluation of the effectiveness of rehabilitation, and
observations of the factors which predict outcome. Psychological medicine.
1974;4:316-25.
Grubaugh AL, Zinzow HM, Paul L, Egede LE, Frueh BC. Trauma exposure and
posttraumatic stress disorder in adults with severe mental illness: a critical review.
Clinical Psychology Review. 2011;31:883-99.
Guest JF, Cookson RF. Cost of schizophrenia to UK Society. An incidence-based cost-

of-illness model for the first 5 years following diagnosis. Pharmacoeconomics.
1999;15:597-610.
Gulbinat W, Dupont A, Jablensky A, Jensen OM, Marsella A, Nakane Y, et al. Cancer

incidence of schizophrenic patients. Results of record linkage studies in three
countries. The British Journal of Psychiatry. 1992:75-83.
Gutierrez-Maldonado J, Caqueo-Urizar A. Effectiveness of a psycho-educational

intervention for reducing burden in Latin American families of patients with
schizophrenia. Quality of Life Research. 2007:739-47.
Guyatt G, Oxman AD, Akl EA, Kunz R, Vist G, Brozek J, et al. GRADE guidelines: 1.
Introduction—GRADE evidence profiles and summary of findings tables. Journal of
Clinical Epidemiology. 2011;64:383-94.
Hack S, Chow B. Pediatric psychotropic medication compliance: a literature review

and research-based suggestions for improving treatment compliance. Journal of
Child and Adolescent Psychopharmacology. 2001;11:59-67.
Haddad PM, Wieck A. Antipsychotic-induced hyperprolactinaemia: mechanisms,

clinical features and management. Drugs. 2004;64:2291-314.
Halvorson A, Whitter M. Approaches to Recovery-Oriented Systems of Care at the

State and Local Levels: Three Case Studies. Rockville, MD: Center for Substance
Abuse Treatment, Substance Abuse and Mental Health Services Administration;
2009.
Hamilton SH, Revicki DA, Edgell ET, Genduso LA, Tollefson G. Clinical and
economic outcomes of olanzapine compared with haloperidol for schizophrenia.
Results from a randomised clinical trial. Pharmacoeconomics. 1999;15:469-80.

Hampton B, Korr W, Bond G, Mayes J, Havis P. Integration service system approach
to avert homelessness: CSP homeless prevention project for HMI adults. Chicago, IL:
Illinois Department of Mental Health and Developmental Disabilities; 1992.
Hansson L, Jormfeldt H, Svedberg P, Svensson B. Mental health professionals'

attitudes towards people with mental illness: do they differ from attitudes held by
people with mental illness? The International Journal of Social Psychiatry.
2013;59:48-54.
Haro JM, Novick D, Bertsch J, Karagianis J, Dossenbach M, Jones PB. Cross-national

clinical and functional remission rates: Worldwide Schizophrenia Outpatient Health
Outcomes (W-SOHO) study. The British Journal of Psychiatry. 2011;199:194-201.
Haro JM, Suarez D, Novick D, Brown J, Usall J, Naber D, et al. Three-year

antipsychotic effectiveness in the outpatient care of schizophrenia: observational
versus randomized studies results. European Neuropsychopharmacology.
2007;17:235-44.
Harrigan SM, McGorry PD, Krstev H. Does treatment delay in first-episode

psychosis really matter? Psychological medicine. 2003;33:97-110.
Harrington M, Lelliott P, Paton C, Konsolaki M, Sensky T, Okocha C. Variation

between services in polypharmacy and combined high dose of antipsychotic drugs
prescribed for in-patients. Psychiatric Bulletin. 2002;26:418-20.
Harris EC, Barraclough B. Excess mortality of mental disorder. The British Journal of
Psychiatry. 1998;173:11-53.
Harrison-Read P, Lucas B, Tyrer P, Ray J, Shipley K, Simmonds S, et al. Heavy users

of acute psychiatric beds: randomized controlled trial of enhanced community
management in an outer London borough. Psychological medicine. 2002;32:403-16.
Harrison G, Hopper K, Craig T, Laska E, Siegel C, Wanderling J, et al. Recovery from

psychotic illness: a 15- and 25-year international follow-up study. The British Journal
of Psychiatry. 2001;178:506-17.
Harrow M, Grossman LS, Jobe TH, Herbener ES. Do patients with schizophrenia
ever show periods of recovery? A 15-year multi-follow-up study. Schizophrenia
Bulletin. 2005;31:723-34.
Harrow M, Jobe TH, Faull RN. Do all schizophrenia patients need antipsychotic
treatment continuously throughout their lifetime? A 20-year longitudinal study.
Psychological Medicine. 2012;42:2145-55.

Hartman N, Leong GB, Glynn SM, Wilkins JN, Jarvik ME. Transdermal nicotine and
smoking behavior in psychiatric patients. American Journal of Psychiatry.
1991;148:374-5.
Hasselblad V. Meta-analysis of multitreatment studies. Medical Decision Making.

1998;18:37-43.
Hasson-Ohayon I, Roe D, Kravetz S. A radomized controlled trial of the effectiveness

of the illness management and recovery program. Psychiatric Services. 2007;58:1461-
66.
Hastrup LH, Kronborg C, Bertelsen M, Jeppesen P, Jorgensen P, Petersen L, et al.

Cost-effectiveness of early intervention in first-episode psychosis: economic
evaluation of a randomised controlled trial (the OPUS study). The British Journal of
Psychiatry. 2013;202:35-41.
Healey A, Knapp M, Astin J, Beecham J, Kemp R, Kirov G, et al. Cost-effectiveness

evaluation of compliance therapy for people with psychosis. The British Journal of
Psychiatry. 1998;172:420-4.
Health and Social Care Information Centre. NHS Staff Earnings Estimates June 2012
- Based on Payments to Staff in the NHS in England from January to March 2012.
Leeds: Health and Social Care Information Centre; 2012.
Healthcare Commission. Count Me In 2008: Results of the 2008 National Census of

Inpatients in Mental Health and Learning Disability Services in England and Wales.
London: Commission for Healthcare Audit and Inspection; 2008.
Healy D. The Creation of Psychopharmacology. Cambridge, MA: Harvard

University Press; 2002.
Hearing Voices Network. Assisting people who hear voices. 2003; Available from:
http://www.hearing-voices.org
Heeg B, Antunes J, Figueira M, Jara J, Marques Teixeira J, Palha A, et al. Cost-

effectiveness and budget impact of long-acting risperidone in Portugal: a modeling
exercise. Current Medical Research and Opinion. 2008;24:349-58.
Heeg B, Buskens E, Knapp M, van Aalst G, Dries PJ, de Haan L, et al. Modelling the
treated course of schizophrenia: development of a discrete event simulation model.
Pharmacoeconomics. 2005;23(Suppl. 1):17-33.
Heinrichs DW, Hanlon TE, Carpenter WT, Jr. The Quality of Life Scale: an
instrument for rating the schizophrenic deficit syndrome. Schizophrenia Bulletin.
1984;10:388-98.

Hemsley DR. A simple (or simplistic?) cognitive model for schizophrenia. Behaviour
Research and Therapy. 1993;31:633-45.
Hennekens CH, Hennekens AR, Hollar D, Casey DE. Schizophrenia and increased
risks of cardiovascular disease. American Heart Journal. 2005;150:1115-21.
Herek GM, Gillis JR, Cogan JC. Psychological sequelae of hate-crime victimization
among lesbian, gay, and bisexual adults. Journal of Consulting and Clinical
Psychology. 1999;67:945.
Herinckx HA, Kinney RF, Clarke GN, Paulson RI. Assertive community treatment
versus usual care in engaging and retaining clients with severe mental illness.
Herman JL, Schatzow E. Recovery and verification of memories of childhood sexual

trauma. Psychoanalytic Psychology. 1987;4:1.
Hersen M, Bellack AS. Social skills training for chronic psychiatric patients: rationale,
research findings, and future directions. Comprehensive Psychiatry. 1976;17:559-80.
Hert M, Correll CU, Bobes J, Cetkovich-Bakmas M, Cohen D, Asai I, et al. Physical

illness in patients with severe mental disorders. I. Prevalence, impact of medications
and disparities in health care. World Psychiatry. 2011;10:52-77.
Heslin M, Howard L, Leese M, McCrone P, Rice C, Jarrett M, et al. Randomized

controlled trial of supported employment in England: 2 Year follow-up of the
Supported Work and Needs (SWAN) study. World Psychiatry. 2011;10:132-37.
Higgins J, Green S, eds. Cochrane Handbook for Systematic Reviews of

Interventions. Version 5.1.0. 2011. Available from www.cochrane-handbook.org.
Oxford: The Cochrane Collaboration.
Higgins JP, Whitehead A. Borrowing strength from external trials in a meta-analysis.

Statistics in Medicine. 1996;15:2733-49.
Himelhoch S, Daumit G. To whom do psychiatrists offer smoking-cessation

counseling? American Journal of Psychiatry. 2003;160:2228-30.
Hippisley-Cox J, Vinogradova Y. Trends in consultation rates in general practice

1995-2008: analysis of the QResearch® database. Report. Leeds: The Information
Centre for Health and Social Care; 2009.
Hirsch SR, Barnes TRE. The clinical treatment of schizophrenia with antipsychotic
medication. In: Hirsch SR, Weinberger DR, eds. Schizophrenia. Oxford: Blackwell;
1995. p. 443–68.

HMSO. Equality Act 2010. London: The Stationary Office; 2010. Available from:
http://www.legislation.gov.uk/ukpga/2010/15/contents.
HMSO. The Mental Capacity Act 2005. London: The Stationery Office; 2005.
Available at: http://www.opsi.gov.uk/acts/acts2005/pdf/ukpga_20050009_en.pdf.
HMSO. The Mental Health Act 2007. London: The Stationery Office; 2007. Available
at: http://www.opsi.gov.uk/acts/acts2007/pdf/ukpga_20070012_en.pdf.
Hoffmann H, Jäckel D, Glauser S, Kupper Z. A randomised controlled trial of the

efficacy of supported employment. Acta Psychiatrica Scandinavica. 2012;125:157-67.
Hogarty GE, Ulrich RF, Mussare F, Aristigueta N. Drug discontinuation among long
term, successfully maintained schizophrenic outpatients. Diseases of the Nervous
System. 1976;37:494-500.
Hollister LE. Clinical differences among phenothiazines in schizophrenics.

Introduction: specific indications for antipsychotics: elusive end of the rainbow.
Advances in Biochemical Psychopharmacology. 1974;9:667-73.
Holloway F, Carson J. Intensive case management for the severely mentally ill.
Controlled trial. The British Journal of Psychiatry. 1998;172:19-22.
Holloway F, Lloyd ES. Inpatient treatment. Thornicroft G, Szmukler G, Mueser KT,

Drake RE, eds. Oxford: Oxford University Press; 2011.
Holt RI, Bushe C, Citrome L. Diabetes and schizophrenia 2005: are we any closer to
understanding the link? Journal of Psychopharmacology. 2005;19:56-65.
Homel P, Casey D, Allison DB. Changes in body mass index for individuals with
and without schizophrenia, 1987-1996. Schizophrenia Research. 2002;55:277-84.
Honer WG, Thornton AE, Chen EY, Chan RC, Wong JO, Bergmann A, et al.
Clozapine alone versus clozapine and risperidone with refractory schizophrenia.
New England Journal of Medicine. 2006;354:472-82.
Hong LE, Thaker GK, McMahon RP, Summerfelt A, Rachbeisel J, Fuller RL, et al.
Effects of moderate-dose treatment with varenicline on neurobiological and
cognitive biomarkers in smokers and nonsmokers with schizophrenia or
schizoaffective disorder. Archives of General Psychiatry. 2011;68:1195-206.
Hor K, Taylor M. Suicide and schizophrenia: a systematic review of rates and risk
factors. Journal of Psychopharmacology. 2010;24:81-90.

Horst WD, Klein MW, Williams D, Werder SF. Extended use of nicotine replacement
therapy to maintain smoking cessation in persons with schizophrenia.
Neuropsychiatric Disease and Treatment. 2005;1:349-55.
Hosalli P, Davis JM. Depot risperidone for schizophrenia. Cochrane Database of

Systematic Reviews. 2003.;4:Art. No. CD004161. DOI: 10.1002/14651858.CD004161.
Hoult J. Home treatment in New South Wales. In: Hall P, Brockington I, eds. The
Closure of Mental Hospitals. London: Gaskell/Royal College of Psychiatrists; 1991.
Hoult J, Reynolds I, Charbonneau-Powis M, Weekes P, Briggs J. Psychiatric hospital

versus community treatment: the results of a randomised trial. Australian and New
Zealand Journal of Psychiatry. 1983;17:160-7.
Howard L. Effectiveness and cost-effectiveness of admissions to women's crisis

houses compared with traditional psychiatric wards: pilot patient-preference
randomised controlled trial. The British Journal of Psychiatry. 2010;197:s32.
Howard LM, Heslin M, Leese M, McCrone P, Rice C, Jarrett M, et al. Supported

employment: randomised controlled trial. The British Journal of Psychiatry.
2010;196:404-11.
Hughes H, Meddings S, Vandrevala T, Holmes S, Hayward M. Carers' experiences

of assertive outreach services: an exploratory study. Journal of Mental Health.
2011;20:70-78.
Intagliata J. Improving the quality of community care for the chronically mentally
disabled: the role of case management. Schizophrenia Bulletin. 1982;8:655-74.
International Early Psychosis Association Writing Group. International clinical

practice guidelines for early psychosis. The British Journal of Psychiatry.
2005;187:s120–s24.
Iqbal MM, Rahman A, Husain Z, Mahmud SZ, Ryan WG, Feldman JM. Clozapine: a
clinical review of adverse effects and management. Annals of Clinical Psychiatry.
2003;15:33-48.
Isaac M. Trends in the development of psychiatric services in India. Psychiatric

Bulletin. 1996;20:43-45.
Jaaskelainen E, Miettunen J, Veijola J, McGrath JJ, Murray GK, Jones PB, et al.
Associations between early development and outcome in schizophrenia- a 35-year
follow-up of the Northern Finland 1966 Birth Cohort. Schizophrenia Research.
2008;99:29-37.

Jablensky A, Sartorius N, Ernberg G, Anker M, Korten A, Cooper JE, et al.
Schizophrenia: manifestations, incidence and course in different cultures. A World
Health Organization ten-country study. Psychological medicine. 1992;20:1-97.
Jackson C, Trower P, Reid I, Smith J, Hall M, Townend M, et al. Improving

psychological adjustment following a first episode of psychosis: a randomised
controlled trial of cognitive therapy to reduce post psychotic trauma symptoms.
Behaviour Research and Therapy. 2009;47:454-62.
Jackson H, McGorry P, Edwards J, Hulbert C, Henry L, Harrigan S, et al. A

controlled trial of cognitively oriented psychotherapy for early psychosis (COPE)
with four-year follow-up readmission data. Psychological medicine. 2005;35:1295-
306.
Jackson HJ, McGorry PD, Killackey E, Bendall S, Allott K, Dudgeon P, et al. Acute-
phase and 1-year follow-up results of a randomized controlled trial of CBT versus
Befriending for first-episode psychosis: the ACE project. Psychological medicine.
2008;38:725-35.
Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al.
Assessing the quality of reports of randomized clinical trials: is blinding necessary?
Controlled Clinical Trials. 1996;17:1-12.
Jankovic J, Yeeles K, Katsakou C, Amos T, Morriss R, Rose D, et al. Family

caregivers' experiences of involuntary psychiatric hospital admissions of their
relatives-a qualitative study. PLoS ONE. 2011;6:e25425.
Janssen I, Krabbendam L, Bak M, Hanssen M, Vollebergh W, Graaf Rd, et al.

Childhood abuse as a risk factor for psychotic experiences. Acta Psychiatrica
Scandinavica. 2004;109:38-45.
Jeppesen P, Petersen L, Thorup A, Abel MB, Oehlenschlaeger J, Christensen T, et al.

Integrated treatment of first-episode psychosis: effect of treatment on family burden:
OPUS trial. The British Journal of Psychiatry. 2005;48:s85-s90.
Jerrell JM. Toward managed care for persons with severe mental illness: implications
from a cost-effectiveness study. Health Affairs. 1995;14:197-207.
Jerrell JM. Cost-effectiveness of risperidone, olanzapine, and conventional

antipsychotic medications. Schizophrenia Bulletin. 2002;28:589-605.
Jeste DV, Gladsjo JA, Lindamer LA, Lacro JP. Medical co-morbidity in
schizophrenia. Schizophrenia Bulletin. 1996;22:413–20.

Ji J, Sundquist K, Ning Y, Kendler KS, Sundquist J, Chen X. Incidence of cancer in
patients with schizophrenia and their first-degree relatives: a population-based
study in Sweden. Schizophrenia Bulletin. 2013;39:527-36.
Joa I, Johannessen JO, Auestad B, Friis S, McGlashan T, Melle I, et al. The key to
reducing duration of untreated first psychosis: information campaigns.
Jobe TH, Harrow M. Long-term outcome of patients with schizophrenia: a review.

Johannessen JO, McGlashan TH, Larsen TK, Horneland M, Joa I, Mardal S, et al.
Early detection strategies for untreated first-episode psychosis. Schizophrenia
Research. 2001;51:39-46.
Johnson S. Crisis resolution and home treatment teams: an evolving model.

Advances in Psychiatric Treatment. 2013;19:115-23.
Johnson S, Gilburt H, Lloyd-Evans B, Osborn DP, Boardman J, Leese M, et al. In-

patient and residential alternatives to standard acute psychiatric wards in England.
Johnson S, Lloyd-Evans B, Howard L, Osborn DPJ, Slade M. Where next with

residential alternatives to admission? The British Journal of Psychiatry. 2010;197:s52-
s54.
Johnson S, Needle J. Crisis resolution teams: rationale and core model. Crisis
Resolution and Home Treatment in Mental Health. 1st edn. Cambridge: Cambridge
University Press; 2008.
Johnson S, Needle J, Bindman J, Thornicroft G. Crisis Resolution and Home

Treatment in Mental Health Cambridge: Cambridge University Press; 2008.
Johnson S, Nolan F, Pilling S, Sandor A, Hoult J, McKenzie N, et al. Randomised

controlled trial of acute mental health care by a crisis resolution team: the north
Islington crisis study. BMJ. 2005;331:599.
Johnston S, Salkeld G, Sanderson K, Issakidis C, Teesson M, Buhrich N. Intensive

case management: a cost-effectiveness analysis. Australian and New Zealand Journal
of Psychiatry. 1998;32:551-9.
Jones B. Olanzapine versus risperidone and haloperidol in the treatment of

schizophrenia. 151st Annual Meeting of the American Psychiatric Association:1998;
Toronto, Ontario.

Jones C, Cormac I, Silveira da Mota Neto JI, Campbell C. Cognitive Behaviour
Therapy for Schizophrenia. The Cochrane Library. 1998.
Jones F, Riazi A. Self-efficacy and self-management after stroke: a systematic review.

Disability and Rehabilitation. 2011;33:797-810.
Jones P. Drama as Therapy: Theatre as Living. London: Routledge; 1996.
Jones PB, Barnes TR, Davies L, Dunn G, Lloyd H, Hayhurst KP, et al. Randomized
controlled trial of the effect on quality of life of second- vs first-generation
antipsychotic drugs in schizophrenia: Cost utility of the latest antipsychotic drugs in
schizophrenia study (CUtLASS 1). Archives of General Psychiatry. 2006;63:1079-87.
Jørgensen P, Nordentoft M, Abel M, Gouliaev G, Jeppesen P, Kassow P. Early

detection and assertive community treatment of young psychotics: the Opus Study
Rationale and design of the trial. Social Psychiatry and Psychiatric Epidemiology.
2000;35:283-87.
Josiassen RC, Joseph A, Kohegyi E, Stokes S, Dadvand M, Paing WW, et al.

Clozapine augmented with risperidone in the treatment of schizophrenia: a
randomized, double-blind, placebo-controlled trial. American Journal of Psychiatry.
2005;162:130-6.
Joy CB, Adams CE, Rice K. Crisis intervention for people with severe mental
illnesses. Cochrane Database of Systematic Reviews. 2002;2: Art.
No.:CD001087.DOI:10.1002/14651858.CD001087.
Kahn RS, Fleischhacker WW, Boter H, Davidson M, Vergouwe Y, Keet IP, et al.
Effectiveness of antipsychotic drugs in first-episode schizophrenia and
schizophreniform disorder: an open randomised clinical trial. The Lancet.
2008;371:1085-97.
Kane J, Honigfeld G, Singer J, Meltzer H. Clozapine for the treatment-resistant

schizophrenic. A double-blind comparison with chlorpromazine. Archives of
General Psychiatry. 1988;45:789-96.
Kane JM. Treatment of schizophrenia. Schizophrenia Bulletin. 1987;13:133-56.
Kane JM. Treatment programme and long-term outcome in chronic schizophrenia.

Acta Psychiatrica Scandinavica Supplementum. 1990;358:151-7.
Kane JM, Marder SR. Psychopharmacologic treatment of schizophrenia.

Kane JM, Marder SR, Schooler NR, Wirshing WC, Umbricht D, Baker RW, et al.
Clozapine and haloperidol in moderately refractory schizophrenia: a 6-month

randomized and double-blind comparison. Archives of General Psychiatry.
2001;58:965-72.
Kaplan K, Salzer MS, Solomon P, Brusilovskiy E, Cousounis P. Internet peer support

for individuals with psychiatric disabilities: a randomized controlled trial. Social
Science and Medicine. 2011;72:54-62.
Kapur S, Mamo D. Half a century of antipsychotics and still a central role for
dopamine D2 receptors. Progress in Neuro-psychopharmacology and Biological
Psychiatry. 2003;27:1081-90.
Kapur S, Remington G. Dopamine D(2) receptors and their role in atypical

antipsychotic action: still necessary and may even be sufficient. Biological
Psychiatry. 2001;50:873-83.
Karow A, Reimer J, Konig HH, Heider D, Bock T, Huber C, et al. Cost-effectiveness

of 12-month therapeutic assertive community treatment as part of integrated care
versus standard care in patients with schizophrenia treated with quetiapine
immediate release (ACCESS trial). Journal of Clinical Psychiatry. 2012;73:e402-e08.
Karunakaran K, Tungaraza TE, Harborne GC. Is clozapine-aripiprazole combination

a useful regime in the management of treatment-resistant schizophrenia? Journal of
Kasckow JW, Twamley E, Mulchahey JJ, Carroll B, Sabai M, Strakowski SM, et al.
Health-related quality of well-being in chronically hospitalized patients with
schizophrenia: comparison with matched outpatients. Psychiatry research.
2001;103:69-78.
Kasper S, Winkler D. Addressing the limitations of the CATIE study. World Journal
of Biological Psychiatry. 2006;7:126-7.
Kavanagh S, Opit L, Knapp M, Beecham J. Schizophrenia: shifting the balance of

care. Social Psychiatry and Psychiatric Epidemiology. 1995;30:206-12.
Kelly DL, McMahon RP, Weiner E, Boggs DL, Dickinson D, Conley RR, et al. Lack of
beneficial galantamine effect for smoking behavior: a double-blind randomized trial
in people with schizophrenia. Schizophrenia Research. 2008;103:161-68.
Kemp R, Hayward P, Applewhaite G, Everitt B, David A. Compliance therapy in

psychotic patients: randomised controlled trial. BMJ. 1996;312:345-9.
Kemp R, Kirov G, Everitt B, Hayward P, David A. Randomised controlled trial of

compliance therapy. 18-month follow-up. The British Journal of Psychiatry.
1998;172:413-9.

Kemp V. Use of ‘chronic disease self-management strategies’ in mental healthcare.
Current Opinion in Psychiatry. 2011;24:144-48.
Kendall T. The rise and fall of the atypical antipsychotics. The British Journal of
Psychiatry. 2011;199:266-68.
Kendrick T, Burns T, Freeling P, Sibbald B. Provision of care to general practice

patients with disabling long-term mental illness: a survey in 16 practices. British
Journal of General Practice. 1994;44:301-5.
Kendrick T, Sibbald B, Burns T, Freeling P. Role of general practitioners in care of

long term mentally ill patients. BMJ. 1991;302:508-10.
Keshavan MS, Nasrallah HA, Tandon R. Schizophrenia, "Just the Facts" 6. Moving
ahead with the schizophrenia concept: from the elephant to the mouse.
Kessler RC, Amminger GP, Aguilar-Gaxiola S, Alonso J, Lee S, Ustun TB. Age of
onset of mental disorders: a review of recent literature. Current Opinion in
Psychiatry. 2007;20:359-64.
Khan I, Pillay K. Users' attitudes towards home and hospital treatment: a

comparative study between South Asian and white residents of the British Isles.
Journal of Psychiatric and Mental Health Nursing. 2003;10:137-46.
Killackey E, Jackson HJ, McGorry PD. Vocational intervention in first-episode

psychosis: individual placement and support v. treatment as usual. The British
Journal of Psychiatry. 2008;193:114-20.
Killaspy H. From the asylum to community care: learning from experience. BMJ.
2006;79-80:245-58.
Killaspy H, Bebbington P, Blizard R, Johnson S, Nolan F, Pilling S, et al. The REACT

study: Randomised evaluation of assertive community treatment in north London.
BMJ. 2006;332:815-18.
Killaspy H, Marston L, Omar RZ, Green N, Harrison I, Lean M, et al. Service quality
and clinical outcomes: an example from mental health rehabilitation services in
England. The British Journal of Psychiatry. 2013;202:28-34.
Kingdon DG, Kinoshita Y, Naeem F, Swelam M, Hansen L, Vincent S, et al.

Schizophrenia can and should be renamed. BMJ. 2007;334:221-2.
Kinon BJ, Kane JM, Johns C, Perovich R, Ismi M, Koreen A, et al. Treatment of
neuroleptic-resistant schizophrenia relapse. Psychopharmacology Bulletin.
1993;29:309-14.

Kirkbride JB, Errazuriz A, Croudace TJ, Morgan C, Jackson D, Boydell J, et al.
Incidence of schizophrenia and other psychoses in England, 1950-2009: a systematic
review and meta-analyses. PLoS One. 2012;7:e31660.
Kirkbride JB, Fearon P, Morgan C, Dazzan P, Morgan K, Tarrant J, et al.

Heterogeneity in incidence rates of schizophrenia and other psychotic syndromes:
findings from the 3-center AeSOP study. Archive of General Psychiatry. 2006;63:250-
8.
Kisely S, Crowe E, Lawrence D. Cancer-related mortality in people with mental

illness. JAMA Psychiatry. 2013;70:209-17.
Kissling W. The current unsatisfactory state of relapse prevention in schizophrenic

psychoses-suggestions for improvement. Clinical Neuropharmacology.
1991;14(Suppl:2):S33-44.
Kleinman A, Benson P. Anthropology in the clinic: the problem of cultural

competency and how to fix it. PLoS medicine. 2006;3:e294.
Kline MN, Hoisington V. Placing the psychiatrically disabled - a look at work values.
Rehabilitation Counseling Bulletin. 1981;24:366-69.
Klosterkotter J, Hellmich M, Steinmeyer EM, Schultze-Lutter F. Diagnosing

schizophrenia in the initial prodromal phase. Archives of General Psychiatry.
2001;58:158-64.
Knapp M, Chisholm D, Leese M, Amaddeo F, Tansella M, Schene A, et al.

Comparing patterns and costs of schizophrenia care in five European countries: the
EPSILON study. European psychiatric services: inputs linked to outcome domains
and needs. Acta Psychiatrica Scandinavica. 2002;105:42-54.
Knapp M, King D, Pugner K, Lapuerta P. Non-adherence to antipsychotic

medication regimens: associations with resource use and costs. British Journal
Psychiatry. 2004;184:509-16.
Knapp M, Patel A, Curran C, Latimer E, Catty J, Becker T, et al. Supported

employment: cost-effectiveness across six European sites. World Psychiatry.
2013;12:60-8.
Knapp M, Windmeijer F, Brown J, Kontodimas S, Tzivelekis S, Haro JM, et al. Cost-

utility analysis of treatment with olanzapine compared with other antipsychotic
treatments in patients with schizophrenia in the pan-European SOHO study.

Knudson B, Coyle A. Parents' experiences of caring for sons and daughters with
schizophrenia: a qualitative analysis of coping. European Journal of Psychotherapy,
Counselling and Health. 2002;5:169-83.
König HH, Roick C, Angermeyer MC. Validity of the EQ-5D in assessing and
valuing health status in patients with schizophrenic, schizotypal or delusional
disorders. European Psychiatry. 2007;22:177-87.
Kontaxakis VP, Ferentinos PP, Havaki-Kontaxaki BJ, Roukas DK. Randomized

controlled augmentation trials in clozapine-resistant schizophrenic patients: a critical
review. European Psychiatry. 2005;20:409-15.
Kontaxakis VP, Havaki-Kontaxaki BJ, Stamouli SS, Christodoulou GN. Toxic

interaction between risperidone and clozapine: a case report. Progress in Neuro-
Psychopharmacology and Biological Psychiatry. 2002;26:407-9.
Koolaee AK, Etemadi A. The outcome of two family interventions for the mothers of
schizophrenia patients in Iran. The International Journal of Social Psychiatry.
2009;56:634-46. E-pub 4 September 2010.
Kopelowicz A. Adapting social skills training for Latinos with schizophrenia.

International Review of Psychiatry. 1998;10:47-50.
Kopelowicz A, Liberman RP, Wallace CJ, Aguirre F, Mintz J. Differential

performance of job skills in schizophrenia: an experimental analysis. Journal of
Rehabilitation. 2006;72:31-39.
Kopelowicz A, Wallace CJ, Zarate R. Teaching psychiatric inpatients to re-enter the

community: a brief method of improving the continuity of care. Psychiatric Services.
1998;49:1313-6.
Koro CE, Fedder DO, L'Italien GJ, Weiss SS, Magder LS, Kreyenbuhl J, et al.
Assessment of independent effect of olanzapine and risperidone on risk of diabetes
among patients with schizophrenia: population based nested case-control study.
BMJ. 2002;325:243.
Krabbendam L, van Os J. Affective processes in the onset and persistence of

psychosis. European Archives of Psychiatry and Clinical Neuroscience. 2005;255:185-
9.
Kreyenbuhl JA, Valenstein M, McCarthy JF, Ganoczy D, Blow FC. Long-term

antipsychotic polypharmacy in the VA health system: patient characteristics and
treatment patterns. Psychiatric Services. 2007;58:489-95.

Krstev H, Carbone S, Harrigan SM, Curry C, Elkins K, McGorry PD. Early
intervention in first-episode psychosis--the impact of a community development
campaign. Social Psychiatry and Psychiatric Epidemiology. 2004;39:711-9.
Kuipers E, Fowler D, Garety P, Chisholm D, Freeman D, Dunn G, et al. London-East

Anglia randomised controlled trial of cognitive-behavioural therapy for psychosis.
III: follow-up and economic evaluation at 18 months. The British Journal of
Psychiatry. 1998;173:61-68.
Kuipers E, Holloway F, Rabe HS, Tennakoon L, Croydon Outreach and Assertive

Support Team. An RCT of early intervention in psychosis: Croydon Outreach and
Assertive Support Team (COAST). Social Psychiatry and Psychiatric Epidemiology.
2004;39:358-63.
Kuipers L, Bebbington P. Working in Partnership: Clinicians and Carers in the

Management of Longstanding Mental Illness. Oxford: Heinemann Medical Books;
1990.
Kuldau JM, Dirks SJ. Controlled evaluation of a hospital-originated community

transitional system. Archives of General Psychiatry. 1977;34:1331-40.
Kumra S, Oberstar JV, Sikich L, Findling RL, McClellan JM, Vinogradov S, et al.
Efficacy and tolerability of second-generation antipsychotics in children and
adolescents with schizophrenia. Schizophrenia Bulletin. 2008;34:60–71.
Kurtz MM, Mueser KT. A meta-analysis of controlled research on social skills

training for schizophrenia. Journal of Consulting and Clinical Psychology.
2008;76:491.
Kwon JS, Choi JS, Bahk WM, Yoon KC, Hyung KC, Chul SY, et al. Weight
management program for treatment-emergent weight gain in olanzapine-treated
patients with schizophrenia or schizoaffective disorder: a 12-week randomized
controlled clinical trial. Journal of Clinical Psychiatry. 2006;67:547-53.
Laird B, Smith B, Dutu G, Mellsop G. Views and experiences of family/whanau

carers of psychiatric service users on diagnosis and classification. The International
Journal of Social Psychiatry. 2010;56:270-79.
Lambert M, Naber D, Schacht A, Wagner T, Hundemer HP, Karow A, et al. Rates

and predictors of remission and recovery during 3 years in 392 never-treated
patients with schizophrenia. Acta Psychiatrica Scandinavica. 2008;118:220-9.
Lamberti JS, Herz MI. Psychotherapy, social skills training, and vocational
rehabilitation in schizophrenia. In: Shriqui C, Nasrallah H, eds. Contemporary Issues
in the Treatment of Schizophrenia. Washington DC: American Psychiatric Press;
1995.

Lang FH, Forbes JF, Murray GD, Johnstone EC. Service provision for people with
schizophrenia. I. Clinical and economic perspective. The British Journal of
Psychiatry. 1997;171:159-64.
Langan J, Mercer SW, Smith DJ. Multimorbidity and mental health: can psychiatry
rise to the challenge? The British Journal of Psychiatry. 2013;202:391-93.
Latimer E. An effective intervention delivered at sub-therapeutic dose becomes an

ineffective intervention. The British Journal of Psychiatry. 2010;196:341-42.
Latimer EA, Lecomte T, Becker DR, Drake RE, Duclos I, Piat M, et al. Generalisability
of the individual placement and support model of supported employment: results of
a Canadian randomised controlled trial. The British Journal of Psychiatry.
2006;189:65-73.
Launois R, Graf von der Schulenberg M, Knapp M, Toumi M. Cost-effectiveness of

sertindole versus olanzapine or haloperidol: a comprehensive model. International
Journal of Psychiatry in Clinical Practice. 1998;2:S79-S86.
Laux G, Heeg B, van Hout BA, Mehnert A. Costs and effects of long-acting
risperidone compared with oral atypical and conventional depot formulations in
Germany. Pharmacoeconomics. 2005;23 (Suppl:1):49-61.
Lawn S. Mental health peer support for hospital avoidance and early discharge: An
Australian example of consumer driven and operated service. Journal of Mental
Health. 2008;17:498-508.
Lawrence D, Hancock KJ, Kisely S. The gap in life expectancy from preventable
physical illness in psychiatric patients in Western Australia: retrospective analysis of
population based registers. BMJ. 2013;346:f2539.
Lawrence D, Kisely S. Inequalities in healthcare provision for people with severe

mental illness. Journal of Psychopharmacology. 2010;24:61-68.
Lawrence DM, Holman CD, Jablensky AV, Hobbs MS. Death rate from ischaemic
heart disease in Western Australian psychiatric patients 1980-1998. The British
Lawrie SM, Martin K, McNeill G, Drife J, Chrystie P, Reid A, et al. General

practitioners' attitudes to psychiatric and medical illness. Psychological medicine.
1998;28:1463-7.
Leavey G, Gulamhussein S, Papadopoulos C, Johnson-Sabine E, Blizard B, King M.

A randomized controlled trial of a brief intervention for families of patients with a
first episode of psychosis. Psychological medicine. 2004;34:423-31.

Lecomte P, De Hert M, van Dijk M, Nuijten M, Nuyts G, Persson U. A 1-year cost-
effectiveness model for the treatment of chronic schizophrenia with acute
exacerbations in Belgium. Value in Health. 2000;3:1-11.
Leff J, Kuipers L, Berkowitz R, Eberlein-Vries R, Sturgeon D. A controlled trial of

social intervention in the families of schizophrenic patients. The British Journal of
Psychiatry. 1982;141:121-34.
Leff MS, Daniel Chisholm, Ray Bell, Catherine Gamble, Julian. Training community
psychiatric nurses in schizophrenia family work: a study of clinical and economic
outcomes for patients and relatives. Journal of Mental Health. 2001;10:189-97.
Lehman AF, Dixon LB, Kernan E, DeForge BR, Postrado LT. A randomized trial of
assertive community treatment for homeless persons with severe mental illness.
Archives of General Psychiatry. 1997;54:1038-43.
Lehman AF, Goldberg R, Dixon LB, McNary S, Postrado L, Hackman A, et al.

Improving employment outcomes for persons with severe mental illnesses. Archives
of General Psychiatry. 2002;59:165-72.
Lehman AF, Steinwachs DM. Patterns of usual care for schizophrenia: initial results
from the Schizophrenia Patient Outcomes Research Team (PORT) Client Survey.
Lehman AF, Steinwachs DM, PORT-Coinvestigators. Patterns of usual care for

schizophrenia: inital survey results from the schizophrenia patient outcomes
research team (PORT) survey. Schizophrenia Bulletin. 1998;24:11-20.
Lenert LA, Sturley AP, Rapaport MH, Chavez S, Mohr PE, Rupnow M. Public
preferences for health states with schizophrenia and a mapping function to estimate
utilities from positive and negative symptom scale scores. Schizophrenia Research.
2004;71:155-65.
Lester H, Birchwood M, Bryan S, England E, Rogers H, Sirvastava N. Development

and implementation of early intervention services for young people with psychosis:
case study. The British Journal of Psychiatry. 2009a;194:446-50.
Lester H, Birchwood M, Freemantle N, Michail M, Tait L. REDIRECT: cluster

randomised controlled trial of GP training in first-episode psychosis. British Journal
of General Practice. 2009b;59:e183-90.
Lester HE, Tritter JQ, Sorohan H. Providing priamry care for people with serious
mental illness: a focus group study. BMJ. 2005:1122-28.

Leucht S, Burkard T, Henderson J, Maj M, Sartorius N. Physical illness and
schizophrenia: a review of the literature. Acta Psychiatrica Scandinavica.
2007;116:317-33.
Leucht S, Cipriani A, Spineli L, Mavridis D, Örey D, Richter F, et al. Comparative

efficacy and tolerability of 15 antipsychotic drugs in schizophrenia: a multiple-
treatments meta-analysis. The Lancet. 2013;382:951-62.
Levine IS, Ligenza LR. In their own voices: families in crisis. A focus group study
families of persons with serious mental illness. Journal of Psychiatric Practice.
2002;8:344-53.
Levitt AJ, Mueser KT, Degenova J, Lorenzo J, Bradford-Watt D, Barbosa A, et al.

Randomized controlled trial of illness management and recovery in multiple-unit
supportive housing. Psychiatric Services. 2009;60:1629-36.
Lewinsohn PM, Hops H, Roberts RE, Seeley JR, Andrews JA. Adolescent
psychopathology: I. Prevalence and incidence of depression and other DSM-III-R
disorders in high school students. Journal of Abnormal Psychology. 1993;102:133-44.
Lewis SW, Barnes TR, Davies L, Murray RM, Dunn G, Hayhurst KP, et al.
Randomised controlled trial of effect of prescription of clozapine versus other
second-generation antipsychotic drugs in resistant schizophrenia. Schizophrenia
Bulletin. 2006a;32:715-23.
Lewis SW, Davies L, Jones PB, Barnes TR, Murray RM, Kerwin R, et al. Randomised
controlled trials of conventional antipsychotic versus new atypical drugs, and new
atypical drugs versus clozapine, in people with schizophrenia responding poorly to,
or intolerant of, current drug treatment. Health Technology Assessment. 2006b;10:1-
165.
Li J, Zhang T, Wang B, Li X. An efficacy analysis of bupropion for smoking cessation

in schizophrenia. Chinese Journal of New Drugs and Clinical Remedies. 2009;28:231-
34.
Liberman RP, Cardin V, Mcgill CW, Falloon IRH, Evans CD. Behavioral family
management of schizophrenia: clinical outcome and costs. Psychiatric Annals.
1987;17:610.
Liberman RP, Kopelowicz A. Training skills for illness self-management in the

rehabilitation of schizophrenia. A family-assisted program for Latinos in California.
Salud Mental. 2009;32:93.
Liberman RP, Wallace CJ, Blackwell G, Eckmaiu T, Kuehnel TG. Skills training for
the seriously mentally ill: modules in the UCLA social and independent living skills

program. In: Ancill R, ed. Schizophrenia: Exploring the Spectrum of Psychosis.
Chichester: Wiley; 1994. p. 35-47.
Liberman RP, Wallace CJ, Blackwell G, Kopelowicz A, Vaccaro JV, Mintz J. Skills
training versus psychosocial occupational therapy for persons with persistent
Lieberman J, Jody D, Geisler S, Alvir J, Loebel A, Szymanski S, et al. Time course and
biologic correlates of treatment response in first-episode schizophrenia. Archives of
Lieberman J, Jody D, Geisler S, Vital-Herne J, Alvir JM, Walsleben J, et al. Treatment

outcome of first episode schizophrenia. Psychopharmacology Bulletin. 1989;25:92-6.
Lieberman JA. Comparative effectiveness of antipsychotic drugs - a commentary on

Cost Utility of the Latest Antipsychotic Drugs in Schizophrenia Study (CUtLASS 1)
and Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE). Archives of
Lieberman JA, Alvir JM, Woerner M, Degreef G, Bilder RM, Ashtari M, et al.
Prospective study of psychobiology in first-episode schizophrenia at Hillside
Hospital. Schizophrenia Bulletin. 1992;18:351-71.
Lieberman JA, Stroup TS, McEvoy JP, Swartz MS, Rosenheck RA, Perkins DO, et al.
Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. New
England Journal of Medicine. 2005;353:1209-23.
Lincoln TM, Arens E, Berger C, Rief W. Can antistigma campaigns be improved? A

test of the impact of biogenetic vs psychosocial causal explanations on implicit and
explicit attitudes to schizophrenia. Schizophrenia Bulletin. 2008;34:984-94.
Lindamer LA, McKibbin C, Norman GJ, Jordan L, Harrison K, Abeyesinhe S, et al.

Assessment of physical activity in middle-aged and older adults with schizophrenia.
Lindenmayer JP, Czobor P, Volavka J, Citrome L, Sheitman B, McEvoy JP, et al.

Changes in glucose and cholesterol levels in patients with schizophrenia treated
with typical or atypical antipsychotics. American Journal of Psychiatry. 2003;160:290-
6.
Lindenmayer JP, Czobor P, Volavka J, Lieberman JA, Citrome L, Sheitman B, et al.

Olanzapine in refractory schizophrenia after failure of typical or atypical
antipsychotic treatment: an open-label switch study. Journal of Clinical Psychiatry.
2002;63:931-35.

Lindenmayer JP, McGurk SR, Mueser KT, Khan A, Wance D, Hoffman L, et al. A
randomized controlled trial of cognitive remediation among inpatients with
persistent mental illness. Psychiatric Services. 2008;59:241-47.
Lindsley OR. Direct measurement and functional definition of vocal hallucinatory

symptoms. The Journal of Nervous and Mental Disease. 1963;136:293-97.
Littrell KH, Hilligoss NM, Kirshner CD, Petty RG, Johnson CG. The effects of an
educational intervention on antipsychotic-induced weight gain. Journal of Nursing
Scholarship. 2003;35:237-41.
Llorca PM, Lancon C, Disdier B, Farisse J, Sapin C, Auquier P. Effectiveness of

clozapine in neuroleptic-resistant schizophrenia: clinical response and plasma
concentrations. Journal of Psychiatry Neuroscience. 2002;27:30-7.
Lloyd-Evans B, Crosby M, Stockton S, Pilling S, Hobbs L, Hinton M, et al. Initiatives

to shorten duration of untreated psychosis: systematic review. The British Journal of
Psychiatry. 2011;198:256-63.
Lloyd-Evans B, Slade M, Jagielska D, Johnson S. Residential alternatives to acute

psychiatric hospital admission: systematic review. The British Journal of Psychiatry.
2009;195:109-17.
Lobban F, Barrowclough C. A Casebook of Family Interventions for Psychosis.

Chichester: Wiley; 2009.
Lobban F, Glentworth D, Chapman L, Wainwright L, Postlethwaite A, Dunn G, et al.

Feasibility of a supported self-management intervention for relatives of people with
recent onset psychosis: REACT study. British Journal of Psychiatry. 2013;203:366-720.
Lobban F, Glentworth D, Haddock G, Wainwright L, Clancy A, Bentley R. The views

of relatives of young people with psychosis on how to design a Relatives Education
And Coping Toolkit (REACT). Journal of Mental Health. 2011;20:567-79.
Loebel AD, Lieberman JA, Alvir JM, Mayerhoff DI, Geisler SH, Szymanski SR.
Duration of psychosis and outcome in first-episode schizophrenia. American Journal
of Psychiatry. 1992;149:1183-8.
Lu G, Ades AE. Combination of direct and indirect evidence in mixed treatment

comparisons. Statistics in Medicine. 2004;23:3105-24.
Lu GB, Ades AE. Assessing evidence inconsistency in mixed treatment comparisons.

Journal of the American Statistical Association. 2006;101:447-59.
Lumsden V, Rajan L. Carer satisfaction with an assertive outreach service. Clinical

Psychology Forum. 2011:11-15.

Lunn DJ, Thomas A, Best N, Spiegelhalter D. WinBUGS - A Bayesian modelling
framework: concepts, structure, and extensibility. Statistics and Computing.
2000;10:325-37.
Lysaker PH, Bond G, Davis LW, Bryson GJ, Bell MD. Enhanced cognitive-behavioral
therapy for vocational rehabilitation in schizophrenia: effects on hope and work.
Journal of Rehabilitation Research and Development. 2005;42:673-82.
Lysaker PH, Davis LW, Bryson GJ, Bell MD. Effects of cognitive behavioral therapy
on work outcomes in vocational rehabilitation for participants with schizophrenia
spectrum disorders. Schizophrenia Research. 2009;107:186-91.
Macias C, Kinney R, Farley OW, Jackson R, Vos B. The role of case management
within a community support system: partnership with psychosocial rehabilitation.
Community Mental Health Journal. 1994;30:323-39.
Mackin P, Bishop D, Watkinson H, Gallagher P, Ferrier IN. Metabolic disease and

cardiovascular risk in people treated with antipsychotics in the community. The
British Journal of Psychiatry. 2007a;191:23-9.
Mackin P, Bishop DR, Watkinson HM. A prospective study of monitoring practices

for metabolic disease in antipsychotic-treated community psychiatric patients. BMC
Psychiatry. 2007b;7:28.
MacMillan JF, Crow TJ, Johnson AL, Johnstone EC. Short-term outcome in trial
entrants and trial eligible patients. The British Journal of Psychiatry. 1986;148:128-33.
Macpherson R, Shepherd G, Thyarappa P. Supported accommodation for people

with severe mental illness: an update. Advances in Psychiatric Treatment.
2012;18:381-91.
Madigan K, Egan P, Brennan D, Hill S, Maguire B, Horgan F, et al. A randomised

controlled trial of carer-focussed multi-family group psychoeducation in bipolar
disorder. European Psychiatry. 2012;27:281-84.
Magliano L, Fadden G, Madianos M, de Almeida JM, Held T, Guarneri M, et al.

Burden on the families of patients with schizophrenia: results of the BIOMED I
study. Social Psychiatry and Psychiatric Epidemiology. 1998;33:405-12.
Magliano L, Fiorillo A, Malangone C, De Rosa C, Maj M. Patient functioning and

family burden in a controlled, real-world trial of family psychoeducation for

Malla A, Bechard-Evans L, Joober R, King S, Abadi S. Understanding the
complexities of delay in treatment of psychosis and relevance for early detection
interventions. Schizophrenia Research. 2006;86:S40.
Malla A, Norman R, Scholten D, Manchanda R, McLean T. A community

intervention for early identification of first episode psychosis: impact on duration of
untreated psychosis (DUP) and patient characteristics. Social Psychiatry and
Psychiatric Epidemiology. 2005;40:337-44.
Malmberg L, Fenton M, Rathbone J. Individual psychodynamic psychotherapy and

psychoanalysis for schizophrenia and severe mental illness (Cochrane review).
Cochrane Database of Systematic Reviews. 2001;3: Art. No.
CD001360.DOI:10.1002/14651858.CD001360.
Mangalore R, Knapp M. Cost of schizophrenia in England. Journal of Mental Health

Policy and Economics. 2007;10:23-41.
Mann T. Clinical Guidelines: Using Clinical Guidelines to Improve Patient Care

Within the NHS. London: NHS Executive; 1996.
Manu P, Correll CU, van Winkel R, Wampers M, De Hert M. Prediabetes in patients

treated with antipsychotic drugs. Journal of Clinical Psychiatry. 2012;73:460-6.
Marder SR, Glynn SM, Wirshing WC, Wirshing DA, Ross D, Widmark C, et al.
Maintenance treatment of schizophrenia with risperidone or haloperidol: 2-year
outcomes. American Journal of Psychiatry. 2003;160:1405-12.
Marder SR, Wirshing DA. Maintenance treatment. In: Hirsch SR, Weinberger DR,
eds. Schizophrenia, 2nd edition. Oxford: Blackwell; 2003.
Marder SR, Wirshing WC, Mintz J, McKenzie J, Johnston K, Eckman TA, et al. Two-
year outcome of social skills training and group psychotherapy for outpatients with
Mari J, Streiner D. Family intervention for schizophrenia. The Cochrane Database of

Systematic Reviews. 2000;2: Art. No. CD000088.
DOI:10.1002/14651858.CD000088.pub2.
Marks IM, Connolly J, Muijen M, Audini B, McNamee G, Lawrence RE. Home-based

versus hospital-based care for people with serious mental illness. The British Journal
of Psychiatry. 1994;165:179-94.
Marrone J, Golowka E. If work makes people with mental illness sick, what do
unemployment, poverty and social isolation cause? Psychiatric Rehabilitation
Journal. 1999;23:187-93.

Marshall M. What have we learnt from 40 years of research on intensive case
management? Epidemiologia e Psichiatria Sociale 2008;17:2.
Marshall M, Crowther R, Almaraz-Serrano A, Creed F, Sledge W, Kluiter H, et al.

Systematic reviews of the effectiveness of day care for people with severe mental
disorders: (1) acute day hospital versus admission; (2) vocational rehabilitation; (3)
day hospital versus outpatient care. Health Technology Assessment. 2001;5:1-75.
Marshall M, Crowther R, Sledge WH, Rathbone J, Soares-Weiser K. Day hospital

versus admission for acute psychiatric disorders. Cochrane Database of Systematic
Reviews. 2011;12: Art. No.CD004026. DOI:1002/14651858.CD004026.pub2.
Marshall M, Gray A, Lockwood A, Green R. Case management for people with

severe mental disorders. Cochrane Database of Systematic Reviews. 2000;12: Art.
No.CD000050. DOI:1002/14651858.CD000050.pub2.
Marshall M, Lewis S, Lockwood A, Drake R, Jones P, Croudace T. Association

between duration of untreated psychosis and outcome in cohorts of first-episode
patients: a systematic review. Archives of General Psychiatry. 2005;62:975-83.
Marshall M, Lockwood A. Assertive community treatment for people with severe

mental disorders – a systematic review. 2000;12: Art. No.CD001089.
DOI:1002/14651858.CD001089.pub2.
Marshall M, Lockwood A. Assertive community treatment for people with severe

mental disorders. Cochrane Database of Systematic Reviews. 2002;2: Art.
No.CD001089. DOI:1002/14651858.CD001089.pub2.
Marshall M, Lockwood A, Gath D. Social services case-management for long-term

mental disorders: a randomised controlled trial. The Lancet. 1995;345:409-12.
Marwaha S, Johnson S. Schizophrenia and employment - a review. Social Psychiatry

and Psychiatric Epidemiology. 2004;39:337-49.
Mauri M, Simoncini M, Castrogiovanni S, Iovieno N, Cecconi D, Dell'Agnello G, et

al. A psychoeducational program for weight loss in patients who have experienced
weight gain during antipsychotic treatment with olanzapine. Pharmacopsychiatry.
2008;41:17-23.
May PRA. Treatment of Schizophrenia: A Comparative Study of Five Treatment

Methods. New York: Science House; 1968
McAuliffe D, Andriske L, Moller E, O'Brien M, Breslin P, Hickey P. 'Who cares?' An

exploratory study of carer needs in adult mental health. Australian e-Journal for the
Advancement of Mental Health. 2009;8:1-12.

McCabe R, Priebe S. The therapeutic relationship in the treatment of severe mental
illness: a review of methods and findings. The International Journal of Social
Psychiatry. 2004;50:115-28.
McCann TV, Lubman DI, Clark E. First-time primary caregivers' experience

accessing first-episode psychosis services. Early Intervention in Psychiatry.
2011;5:156-62.
McCann TV, Lubman DI, Clark E. Primary caregivers' satisfaction with clinicians'
response to them as informal carers of young people with first-episode psychosis: a
qualitative study. Journal of Clinical Nursing. 2012a;21:224-31.
McCann TV, Lubman DI, Cotton SM, Murphy B, Crisp K, Catania L, et al. A
randomized controlled trial of bibliotherapy for carers of young people with first-
episode psychosis. Schizophrenia Bulletin. 2012b.
McCarthy RH. Seizures following smoking cessation in a clozapine responder.

Mccarthy RH, Terkelsen KG. Risperidone augmentation of clozapine.

McCrone P, Beecham J, Knapp M. Community psychiatric nurse teams: cost-

effectiveness of intensive support versus generic care. The British Journal of
Psychiatry. 1994;165:218-21.
McCrone P, Craig TK, Power P, Garety PA. Cost-effectiveness of an early

intervention service for people with psychosis. The British Journal of Psychiatry.
2010;196:377-82.
McCrone P, Johnson S, Nolan F, Pilling S, Sandor A, Hoult J, et al. Economic

evaluation of a crisis resolution service: a randomised controlled trial. Epidemiologia
e Psichiatria Sociale. 2009a;18:54-8.
McCrone P, Johnson S, Nolan F, Pilling S, Sandor A, Hoult J, et al. Impact of a crisis

resolution team on service costs in the UK. Psychiatric Bulletin. 2009b;33:17-19.
McCrone P, Killaspy H, Bebbington P, Johnson S, Nolan F, Pilling S, et al. The

REACT study: cost-effectiveness analysis of assertive community treatment in north
London. Psychiatric Services. 2009c;60:908-13.
McCrone P, Knapp M, Dhanasiri S. Economic impact of services for first-episode

psychosis: a decision model approach. Early Intervention in Psychiatry. 2009d;3:266-
73.

McCrone P, Singh SP, Knapp M, Smith J, Clark M, Shiers D. The economic impact of
early intervention in psychosis services for children and adolescents. Early
Intervention in Psychiatry. 2013;7:368-73.
McDonel EC, Bond GR, Salyers M, Fekete D, Chen A, McGrew JH, et al.
Implementing assertive community treatment programs in rural settings.
Administration and Policy in Mental Health. 1997;25:153- 73.
McEvoy JP, Schooler NR, Wilson WH. Predictors of therapeutic response to

haloperidol in acute schizophrenia. Psychopharmacology Bulletin. 1991;27:97-101.
McFarlane WR, Dushay RA, Deakins SM, Stastny P, Lukens EP, Toran J, et al.
Employment outcomes in family-aided assertive community treatment. American
Journal of Orthopsychiatry. 2000;70:203-14.
McFarlane WR, Lukens E, Link B, Dushay R, Deakins SA, Newmark M, et al.

Multiple-family groups and psychoeducation in the treatment of schizophrenia.
Archives of General Psychiatry. 1995;52:679.
McGlashan TH. The Chestnut Lodge follow-up study: II. Long-term outcome of
schizophrenia and the affective disorders. Archives of General Psychiatry.
1984;41:586.
McGlashan TH, Zipursky RB, Perkins D, Addington J, Miller T, Woods SW, et al.
Randomized, double-blind trial of olanzapine versus placebo in patients
prodromally symptomatic for psychosis. American Journal of Psychiatry.
2006;163:790-9.
McGlashan TH, Zipursky RB, Perkins D, Addington J, Miller TJ, Woods SW, et al.
The PRIME North America randomized double-blind clinical trial of olanzapine
versus placebo in patients at risk of being prodromally symptomatic for psychosis. I.
Study rationale and design. Schizophrenia Research. 2003;61:7-18.
McGorry PD, Edwards J, Mihalopoulos C, Harrigan SM, Jackson HJ. EPPIC: an

evolving system of early detection and optimal management. Schizophrenia Bulletin.
1996;22:305-26.
McGorry PD, Hickie IB, Yung AR, Pantelis C, Jackson HJ. Clinical staging of
psychiatric disorders: a heuristic framework for choosing earlier, safer and more
effective interventions. Australian and New Zealand Journal of Psychiatry.
2006;40:616-22.
McGorry PD, Yung AR, Phillips LJ, Yuen HP, Francey S, Cosgrave EM, et al.
Randomized controlled trial of interventions designed to reduce the risk of
progression to first-episode psychosis in a clinical sample with subthreshold
symptoms. Archives of General Psychiatry. 2002;59:921-8.

McGrath J, Saha S, Chant D, Welham J. Schizophrenia: a concise overview of
incidence, prevalence, and mortality. Epidemiologic Reviews. 2008;30:67-76.
McGrath JJ. Variations in the incidence of schizophrenia: data versus dogma.

McGrew JH, Bond GR. Critical ingredients of assertive community treatment:

judgments of the experts. The Journal of Mental Health Administration. 1995;22:113-
25.
McGurk SR, Mueser KT, DeRosa TJ, Wolfe R. Work, recovery, and comorbidity in
schizophrenia: a randomized controlled trial of cognitive remediation. Schizophrenia
Bulletin. 2009;35:319-35.
McGurk SR, Mueser KT, Pascaris A. Cognitive training and supported employment
for persons with severe mental illness: one-year results from a randomized
controlled trial. Schizophrenia Bulletin. 2005;31:898-909.
McGurk SR, Twamley EW, Sitzer DI, McHugo GJ, Mueser KT. A meta-analysis of
cognitive remediation in schizophrenia. American Journal of Psychiatry.
2007;164:1791-802.
McIntosh AM, Conlon L, Lawrie SM, Stanfield AC. Compliance therapy for
schizophrenia (Cochrane review). Cochrane Database of Systematic Reviews. 2006;3:
Art. No. CD003442.pub 2. DOI:10.1002/14651858.CD003442.pub2.
McKenzie K, Bhui K. Institutional racism in mental health care. BMJ. 2007;334:649-50.
McKibbin CL, Patterson TL, Norman G, Patrick K, Jin H, Roesch S, et al. A lifestyle
intervention for older schizophrenia patients with diabetes mellitus: a randomized
controlled trial. Schizophrenia Research. 2006;86:36-44.
Meaney AM, Smith S, Howes OD, O'Brien M, Murray RM, O'Keane V. Effects of
long-term prolactin-raising antipsychotic medication on bone mineral density in
patients with schizophrenia. The British Journal of Psychiatry. 2004;184:503-8.
Meli G, Ottl B, Paladini A, Cataldi L. Prenatal and perinatal risk factors of

schizophrenia. The Journal of Maternal-Fetal and Neonatal Medicine. 2012;25:2559-
63.
Melle I, Larsen TK, Haahr U, Friis S, Johannessen JO, Opjordsmoen S, et al. Reducing
the duration of untreated first-episode psychosis: effects on clinical presentation.

Merson S, Tyrer P, Onyett S, Lack S, Birkett P, Lynch S, et al. Early intervention in
psychiatric emergencies: a controlled clinical trial. The Lancet. 1992;339:1311-14.
Michon HW, van Weeghel J, Kroon H, Schene AH. Person-related predictors of

employment outcomes after participation in psychiatric vocational rehabilitation
programmes-a systematic review. Social Psychiatry and Psychiatric Epidemiology.
2005;40:408-16.
Middleton H, Glover G, Onyett S, Linde K. Crisis resolution/home treatment teams,

gate-keeping and the role of the consultant psychiatrist. Psychiatric Bulletin.
2008;32:378-79.
Mihalopoulos C, Harris M, Henry L, Harrigan S, McGorry P. Is early intervention in

psychosis cost-effective over the long term? Schizophrenia Bulletin. 2009;35:909-18.
Miller B, Messias E, Miettunen J, Alaraisanen A, Jarvelin MR, Koponen H, et al.

Meta-analysis of paternal age and schizophrenia risk in male versus female
offspring. Schizophrenia Bulletin. 2011;37:1039-47.
Miller P, Lawrie S, Hodges A, Clafferty R, Cosway R, Johnstone E. Genetic liability,

illicit drug use, life stress and psychotic symptoms: preliminary findings from the
Edinburgh study of people at high risk for schizophrenia. Social Psychiatry and
Psychiatric Epidemiology. 2001;36:338-42.
Miller WR, Rollnick S. Motivational Interviewing: Preparing People to Change

Addictive Behaviour. New York: Guilford Press; 1991.
Mind. Listening to experience: an independent report into acute and crisis mental
healthcare. London: Mind; 2011; Available from:
http://www.mind.org.uk/assets/0001/5921/Listening_to_experience_web.pdf.
Moffat J, Sass B, Mckenzie K, Bhui K. Improving pathways into mental health care
for black and ethnic minority groups: a systematic review of the grey literature.
Mohan R, McCrone P, Szmukler G, Micali N, Afuwape S, Thornicroft G. Ethnic

differences in mental health service use among patients with psychotic disorders.
Social Psychiatry and Psychiatric Epidemiology. 2006;41:771-76.
Möller HJ. Do effectiveness ("real world") studies on antipsychotics tell us the real
truth? European Archives of Psychiatry and Clinical Neuroscience. 2008;258:257-70.
Möller HJ, van Zerssen D. Course and outcome of schizophrenia. In: Hirsch. S,
Weinberger D, eds. Schizophrenia Oxford: Blackwell; 1995. p. 106–27.

Moncrieff J. A critique of the dopamine hypothesis of schizophrenia and psychosis.
Harvard Review of Psychiatry. 2009;17:214-25.
Moore TH, Zammit S, Lingford-Hughes A, Barnes TR, Jones PB, Burke M, et al.
Cannabis use and risk of psychotic or affective mental health outcomes: a systematic
review. The Lancet. 2007;370:319-28.
Morant N, Lloyd-Evans B, Gilburt H, Slade M, Osborn D, Johnson S. Implementing

successful residential alternatives to acute in-patient psychiatric services: lessons
from a multi-centre study of alternatives in England. Epidemiologia e Psichiatria
Sociale. 2012;21:175-85.
Morgan C, Dazzan P, Morgan K, Jones P, Harrison G, Leff J, et al. First episode

psychosis and ethnicity: initial findings from the AESOP study. World Psychiatry.
2006;5:40.
Morgan C, Kirkbride J, Leff J, Craig T, Hutchinson G, McKenzie K, et al. Parental

separation, loss and psychosis in different ethnic groups: a case-control study.
Morrato EH, Newcomer JW, Kamat S, Baser O, Harnett J, Cuffel B. Metabolic

screening after the American Diabetes Association's consensus statement on
antipsychotic drugs and diabetes. Diabetes Care. 2009;32:1037-42.
Morrison AP, Byrne R, Bentall RP. DSM-5 and the ‘psychosis risk syndrome’: whose
best interests would it serve? Psychosis. 2010;2:96-99.
Morrison AP, French P, Parker S, Roberts M, Stevens H, Bentall RP, et al. Three-year
follow-up of a randomized controlled trial of cognitive therapy for the prevention of
psychosis in people at ultrahigh risk. Schizophrenia Bulletin. 2007;33:682-7.
Morrison AP, French P, Walford L, Lewis SW, Kilcommons A, Green J, et al.

Cognitive therapy for the prevention of psychosis in people at ultra-high risk:
randomised controlled trial. The British Journal of Psychiatry. 2004;185:291-97.
Morrison AP, Hutton P, Shiers D, Turkington D. Antipsychotics: is it time to

introduce patient choice? The British Journal of Psychiatry. 2012a;201:83-84.
Morrison AP, Hutton P, Wardle M, Spencer H, Barratt S, Brabban A, et al. Cognitive

therapy for people with a schizophrenia spectrum diagnosis not taking antipsychotic
medication: an exploratory trial. Psychological Medicine. 2012b;42:1049-56.
Morse G, Calsyn R, Allen G, Tempelhoff B, Smith R. Experimental comparison of the

effects of three treatment programs for homeless mentally ill people. Hospital and

Morse GA, Calsyn RJ, Klinkenberg WD, Helminiak TW, Wolff N, Drake RE, et al.
Treating homeless clients with severe mental illness and substance use disorders:
costs and outcomes. Community Mental Health Journal. 2006;42:377-404.
Mossaheb N, Sacher J, Wiesegger G, Klein N, Spindelegger CJ, Asenbaum S, et al.

Haloperidol in combination with clozapine in treatment-refractory patients with
schizophrenia. European Neuropsychopharmacology. 2006;16:S416-S16.
Mueser KT, Aalto S, Becker DR, Ogden JS, Wolfe RS, Schiavo D, et al. The
effectiveness of skills training for improving outcomes in supported employment.
Mueser KT, Clark RE, Haines M, Drake RE, Bond GR, Becker DR, et al. The Hartford
study of supported employment for severe mental illness: employment and
nonvocational outcomes. 155th Annual Meeting of the American Psychiatric
Association:2002a; Philadelphia, PA.
Mueser KT, Clark RE, Haines M, Drake RE, McHugo GJ, Bond GR, et al. The
Hartford study of supported employment for persons with severe mental illness.
Journal of Consulting and Clinical Psychology. 2004;72:479-90.
Mueser KT, Corrigan PW, Hilton DW, Tanzman B, Schaub A, Gingerich S, et al.
Illness management and recovery: a review of the research. Psychiatric Services.
2002b;53:1272-84.
Mueser KT, Gingerich S. Illness self-management programmes. In: Thornicroft G,

Szmukler G, Mueser K, Drake R, eds. Oxford Textbook of Community Mental
Health. Oxford: Oxford University Press; 2011.
Muijen M, Marks I, Connolly J, Audini B. Home based care and standard hospital
care for patients with severe mental illness: a randomised controlled trial. BMJ.
1992;304:749-54.
Muller-Clemm WJ. Halting the ’revolving door’ of serious mentaliIllness: evaluating

an assertive case management programme (deinstitutionalization, community
mental health). Victoria, Canada: University of Victoria; 1996.
Murphy S, Irving CB, Adams CE, Driver R. Crisis intervention for people with
severe mental illnesses. Cochrane Database of Systematic Reviews. 2012;5: Art.
No.CD001087.DOI:10.1002/14651858.CD00087.
Murray CJ, Richards MA, Newton JN, Fenton KA, Anderson HR, Atkinson C, et al.
UK health performance: findings of the global burden of disease study 2010. The
Lancet. 2013;381:997-1020.

Murray CJ, Vos T, Lozano R, Naghavi M, Flaxman AD, Michaud C, et al. Disability-
adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990-2010: a
systematic analysis for the global burden of disease study 2010. The Lancet.
2012;380:2197-223.
Mustafa RA, Santesso N, Brozek J, Akl EA, Walter SD, Norman G, et al. The GRADE
approach is reproducible in assessing the quality of evidence of quantitative
evidence syntheses. Journal of Clinical Epidemiology. 2013;66:736-42.
Myles N, Newall HD, Curtis J, Nielssen O, Shiers D, Large M. Tobacco use before, at,
and after first-episode psychosis: a systematic meta-analysis. Journal of Clinical
Psychiatry. 2012;73:468-75.
Nagel T, Robinson G, Condon J, Trauer T. Approach to treatment of mental illness

and substance dependence in remote Indigenous communities: results of a mixed
methods study. The Australian Journal of Rural Health. 2009;17:174-82.
Nasrallah H. A review of the effect of atypical antipsychotics on weight.

Psychoneuroendocrinology. 2003;28(Suppl:1):83-96.
Nasrallah H. Folie en masse! It’s so tempting to drink the Kool-Aid (from the Editor).
Current Psychiatry. 2011;10.
Nasrallah HA. Atypical antipsychotic-induced metabolic side effects: insights from

receptor-binding profiles. Molecular Psychiatry. 2008;13:27-35.
Nasrallah HA, Meyer JM, Goff DC, McEvoy JP, Davis SM, Stroup TS, et al. Low rates
of treatment for hypertension, dyslipidemia and diabetes in schizophrenia: data
from the CATIE schizophrenia trial sample at baseline. Schizophrenia Research.
2006;86:15-22.
National Institute for Mental Health in England. NIMHE Inside Outside. Improving
Mental Health Services for Black and Minority Ethnic Communities. London:
Department of Health; 2003; Available from: www.nimhe.org.uk/
downloads/inside_outside.pdf.
National Institute for Mental Health in England. NIMHE Guiding Statement on

Recovery. London: National Institute for Mental Health in England; 2005.
Nayor C, Bell B. Mental health and the productivity challenge: improving quality
and value for money. Report. London: The King’s Fund and Centre for Mental
Health; 2010.
Nazareth I, King M, Haines A, Rangel L, Myers S. Accuracy of diagnosis of

psychosis on general practice computer system. BMJ. 1993;307:32-4.

Nazareth ID, King MB. Controlled evaluation of management of schizophrenia in
one general practice: a pilot study. Family Practice. 1992;9:171-2.
NCCMH. Schizophrenia: Full National Clinical Guideline on Core Interventions in

Primary and Secondary Care. Leicester and London: The British Psychological
Society and the Royal College of Psychiatrists; 2003.
NCCMH. Bipolar Disorder: the Management of Bipolar Disorder in Adults, Children

and Adolescents, in Primary and Secondary Care. Clinical guideline No. 38.
Leicester & London: The British Psychological Society and Gaskell; 2006 [full
guideline].
NCCMH. Schizophrenia: Core Interventions in the Treatment and Management of

Schizophrenia in Adults in Primary and Secondary Care. Updated edition. Clinical
guideline No. 82. Leicester and London: The British Psychological Society and the
Royal College of Psychiatrists; 2010 [full guideline].
NCCMH. Service User Experience in Adult Mental Health: Improving the

Experience of Care for People Using Adult NHS Mental Health Services. Clinical
guideline No. 136. London: RCPsych Publications; 2012 [full guideline].
NCCMH. Psychosis and Schizophrenia in Children and Young People: Recognition

and Management. Clinical guideline No. 155. Leicester and London: The British
Psychological Society and the Royal College of Psychiatrists; 2013 [full guideline].
Neelam K, Garg D, Marshall M. A systematic review and meta-analysis of

neurological soft signs in relatives of people with schizophrenia. BMC Psychiatry.
2011;11:139.
Nelson MT, Seal ML, Pantelis C, Phillips LJ. Evidence of a dimensional relationship
between schizotypy and schizophrenia: a systematic review. Neuroscience and
Biobehavioral Reviews. 2013;37:317-27.
Newcomer JW. Antipsychotic medications: metabolic and cardiovascular risk.

Journal of Clinical Psychiatry. 2007;68(Suppl:4):8-13.
Newcomer JW, Haupt DW. The metabolic effects of antipsychotic medications.

Newcomer JW, Wieden PJ, Buchanan RW. Switching antipsychotic medications to

reduce adverse event burden in schizophrenia: establishing evidence-based practice.
Journal of Clinical Psychiatry. 2013;74:1108-20.
NHS Business Services Authority, Prescription Pricing Division. Electronic Drug

Tariff for England and Wales, June 2008. Compiled on behalf of the Department of
Health. 2008 [updated December 2008].

NHS Employers. Pay Circular (AforC) 1/2006. Pay and Conditions for NHS Staff
Covered by the Agenda for Change Agreement. London: NHS Employers; 2006.
NHS Employers. Quality and Outcomes Framework guidance for GMS contract
2011/12. London: NHS Employers; 2011.
NHS The Information Centre. Average Daily Quantity Values. London: The NHS
Information Centre, Prescribing Support Unit, NHS; 2008a; Available from:
http://www.ic.nhs.uk/webfiles/Services/PSU/adqs_2007_08.pdf.
NHS The Information Centre. Hospital Episode Statistics 2006–07. London: The
NHS Information Centre; 2008b; Available from: http://www.hesonline.nhs.uk.
NHS The Information Centre. Prescription Cost Analysis England 2007. London:
The NHS Information Centre, Prescribing Support Unit, NHS; 2008c; Available from:
http://www.ic.nhs.uk/pubs/prescostanalysis2007.
NICE. Guidance on the Use of Newer (Atypical) Antipsychotic Drugs for the
Treatment of Schizophrenia: Technology Appraisal. London: NICE; 2002a.
NICE. Schizophrenia: Core Interventions in the Treatment and Management of

Schizophrenia in Primary and Secondary Care. NICE clinical guideline 1. London:
NICE; 2002b.
NICE. Bipolar Disorder: The Management of Bipolar Disorder in Adults, Children

and Adolescents, in Primary and Secondary Care. Clinical guideline no. 38. London:
National Institute for Health and Clinical Excellence; 2006a. Available from:
www.nice.org.uk/CG38.
NICE. Obesity: Guidance on the Prevention, Identification, Assessment and

Management of Overweight and Obesity in Adults and Children. NICE clinical
guideline 43. London: NICE; 2006b.[cited Chapter 7]:Available from:
NICE. The Guidelines Manual. London: NICE; 2007.
NICE. Guide to the Methods of Technology Appraisal. London: NICE; 2008a.
NICE. Social Value Judgements. Principles for the Development of NICE Guidance.
2nd edition. London: NICE; 2008b.
NICE. Depression in Adults: The Treatment and Management of Depression in

Adults. NICE clinical guideline no. 90. London: National Institute of Health and
Clinical Excellence; 2009a. Available from: www.nice.org.uk/CG90.

NICE. Management of Long-term Sickness and Incapacity for Work. Public health
guidance 19. London: NICE; 2009b.
NICE. Promoting Mental Wellbeing Through Productive and Healthy Working

Conditions: Guidance for Employers. Public health guidance 22. London: NICE;
2009c.
NICE. Schizophrenia: Core Interventions in the Treatment and Management of

Schizophrenia in Adults in Primary and Secondary Care (Update). NICE clinical
guideline 82. London: NICE; 2009d. Available from: www.nice.org.uk/CG82.
NICE. Service user experience in adult mental health: improving the experience of
care for people using adult mental health service. Clinical guideline CG136 2011.
Available from: http://guidance.nice.org.uk/CG136.
NICE. Autism: Recognition, Referral, Diagnosis and Management of Adults on the

Autism Spectrum. Clinical guideline no. 142. London: NICE; 2012a. Available from:
NICE. The Guidelines Manual. London: NICE; 2012b; Available from:

http://publications.nice.org.uk/pmg6.
NICE. Preventing Type 2 Diabetes: Risk Identification and Interventions for

Individuals at High Risk. NICE Public health guidance 38. 2012c. Available from:
http://www.nice.org.uk/PH38.
NICE. Psychosis and Schizophrenia in Children and Young People: Recognition and
Management. NICE clinical guideline no. 155. London: NICE; 2013a. Available from:
NICE. Smoking Cessation Services. NICE public health guidance no. 10. London:
NICE; 2013b.
Nicholls CJ, Hale AS, Freemantle N. Cost-effectiveness of amisulpride compared

with risperidone in patients with schizophrenia. Journal of Drug Assessment.
2003;6:79-89.
Nicholls E, Pernice R. Perceptions of the relationship between mental health

professionals and family caregivers: has there been any change? Issues in Mental
Health Nursing. 2009;30:474-81.
Nordby K, Kjonsberg K, Hummelvoll JK. Relatives of persons with recently

discovered serious mental illness: In need of support to become resource persons in
treatment and recovery. Journal of Psychiatric and Mental Health Nursing.
2010;17:304-11.

Norman R, Malla A, Verdi M, Hassall L, Fazekas C. Understanding delay in
treatment for first episode psychosis. Psychological Medicine. 2006;34:255-66.
Nose M, Barbui C, Tansella M. How often do patients with psychosis fail to adhere
to treatment programmes? A systematic review. Psychological medicine.
2003;33:1149-60.
Nuechterlein KH. Vulnerability models for schizophrenia: state of the art. In Search
for the Causes of Schizophrenia. In: Hafner H, Gattaz WF, Janzarik W, eds.
Heidelberg: Springer; 1987. p. 297–316.
Nuechterlein KH, Barch DM, Gold JM, Goldberg TE, Green MF, Heaton RK.
Identification of separable cognitive factors in schizophrenia. Schizophrenia
Research. 2004;72:29-39.
Office for National Statistics. Population Trends 131. Deaths: Age and Sex, Numbers
and Rates, 1976 Onwards (England and Wales). London: Office for National
Statistics; 2008; Available from:
http://www.statistics.gov.uk/STATBASE/ssdataset.asp?vlnk=9552&More=Y.
Office of the Deputy Prime Minister. Mental Health and Social Exclusion. Report.
London: Office of the Deputy Prime Minister Publications; 2004.
Oh PI, Lanctôt KL, Mittmann N, Iskedjian M, Einarson TR. Cost-utility of

risperidone compared with standard conventional antipsychotics in chronic
schizophrenia. Journal of Medical Economics. 2001;4:137-56.
Okpaku SO, Anderson KH. The effectiveness of a multidisciplinary case

management. Psychiatric Rehabilitation Journal. 1997;20:34.
Oltmanns TF, Neale JM. Schizophrenic performance when distractors are present:
attentional deficit or differential task difficulty? Journal of Abnormal Psychology.
1975;84:205-9.
Oosthuizen P, Emsley RA, Turner J, Keyter N. Determining the optimal dose of

haloperidol in first-episode psychosis. Journal of Psychopharmacology. 2001;15:251-
5.
Organisation for Economic Co-operation and Development. Mental Health and

Work Project: Organisation for Economic Co-operation and Development; 2011.
Available from:
http://www.oecd.org/document/20/0,3746,en_2649_33933_38887124_1_1_1_1,00.h
tml.
Osborn DP, Levy G, Nazareth I, Petersen I, Islam A, King MB. Relative risk of
cardiovascular and cancer mortality in people with severe mental illness from the

United Kingdom's General Practice Research Database. Archive of General
Psychiatry. 2007a;64:242-49.
Osborn DP, Nazareth I, King MB. Physical activity, dietary habits and Coronary
Heart Disease risk factor knowledge amongst people with severe mental illness: a
cross sectional comparative study in primary care. Social Psychiatry and Psychiatric
Epidemiology. 2007b;42:787-93.
Osborn DP, Nazareth I, Wright CA, King MB. Impact of a nurse-led intervention to
improve screening for cardiovascular risk factors in people with severe mental
illnesses. Phase-two cluster randomised feasibility trial of community mental health
teams. BMC Health Services Research 2010a;10:61.
Osborn DPJ, King MB, Nazareth I. Risk of cardiovascular disease in people with
severe mental illness: a cross sectional comparative study in primary
care. The British Journal of Psychiatry. 2006:271–77.
Osborn DPJ, Lloyd-Evans B, Johnson S, Gilburt H, Byford S, Leese M, et al.

Residential alternatives to acute in-patient care in England: satisfaction, ward
atmosphere and service user experiences. The British Journal of Psychiatry.
2010b;197:S41-S45.
Ozdemir V, Kalow W, Posner P, Collins EJ, Kennedy JL, Tang BK, et al. CYP1A2
activity as measured by a caffeine test predicts clozapine and active metabolite
steady-state concentrationin patients with schizophrenia. Journal of Clinical
Pajonk FG, Wobrock T, Gruber O, Scherk H, Berner D, Kaizl I, et al. Hippocampal

plasticity in response to exercise in schizophrenia. Archives of General Psychiatry.
2010;67:133-43.
Palmer CS, Brunner E, Ruiz-Flores LG, Paez-Agraz F, Revicki DA. A cost-

effectiveness clinical decision analysis model for treatment of schizophrenia.
Archives of Medical Research. 2002;33:572-80.
Palmer CS, Revicki DA, Genduso LA, Hamilton SH, Brown RE. A cost-effectiveness
clinical decision analysis model for schizophrenia. American Journal of Managed
Care. 1998;4:345-55.
Pantelis C, Lambert TJ. Managing patients with 'treatment resistant' schizophrenia.

Medical Journal of Australia. 2003;178 (Suppl.):62-66.
Papadopoulos I, Tilki M, Lees S. Promoting cultural competence in healthcare

through a research-based intervention in the UK. Diversity in Health and Social
Care. 2004;1:107-16.

Pasamanick B, Scarpitti F, Lefton M, Dinitz S, Wernert JJ, McPheerers H. Home vs
hospital care for schizophrenics. Jama: Journal of the American Medical Association.
1964;1:177-87.
Patel MX, David AS. Why aren’t depot antipsychotics prescribed more often and
what can be done about it? Advances in Psychiatric Treatment. 2005;11:203-11.
Paton C, Barnes TR, Cavanagh MR, Taylor D, Lelliott P. High-dose and combination
antipsychotic prescribing in acute adult wards in the UK: the challenges posed by
p.r.n. prescribing. The British Journal of Psychiatry. 2008;192:435-9.
Paton C, Lelliott P, Harrington M, Okocha C, Sensky T, Duffett R. Patterns of

antipsychotic and anticholinergic prescribing for hospital inpatients. Journal of
Paton C, Whittington C, Barnes TR. Augmentation with a second antipsychotic in

patients with schizophrenia who partially respond to clozapine: a meta-analysis.
Journal of Clinical Psychopharmacology. 2007;27:198-204.
Patterson TL, Mausbach BT, McKibbin C, Goldman S, Bucardo J, Jeste DV.

Functional adaptation skills training (FAST): a randomized trial of a psychosocial
intervention for middle-aged and older patients with chronic psychotic disorders.
Patterson TL, McKibbin C, Taylor M, Goldman S, Davila FW, Bucardo J, et al.

Functional adaptation skills training (FAST): a pilot psychosocial intervention study
in middle-aged and older patients with chronic psychotic disorders. American
Journal of Geriatric Psychiatry. 2003;11:17-23.
Paul KI, Moser K. Unemployment impairs mental health: Meta-analyses. Journal of

Vocational Behavior. 2009;74:264-82.
Peasgood T, Roberts J, Tsuchiya A. Incapacity benefit: a health or labour market

phenomenon? Sheffield economic research paper series number: 2006011. Sheffield:
Department of Economics, University of Sheffield; 2006.
Pekkala E, Merinder L. Psychoeducation for schizophrenia. The Cochrane Library.

2002;2: Art. No.CD002831.DOI:1001002/14651858.CD002831.pub 2.
Perkins DO, Gu H, Boteva K, Lieberman JA. Relationship between duration of

untreated psychosis and outcome in first-episode schizophrenia: a critical review
and meta-analysis. American Journal of Psychiatry. 2005;162:1785-804.
Perkins R, Repper J, Rinaldi M, Brown H. Briefing. Recovery colleges: implementing

recovery through organisational change. Centre for Mental Health; NHS
Confederation Mental Health Network. London and Leeds;2012. Available from:

http://www.nhsconfed.org/Documents/ImROC%20Briefing%20Recovery%20Colle
ges.pdf.
Perkins R, Slade M. Recovery in England: transforming statutory services?

Perlick DA, Miklowitz DJ, Lopez N, Chou J, Kalvin C, Adzhiashvili V, et al. Family-
focused treatment for caregivers of patients with bipolar disorder. Bipolar Disorders.
2010;12:627-37.
Petersen L, Jeppesen P, Thorup A, Abel MB, Øhlenschlaeger J, Christensen TØ, et al.

A randomised multicentre trial of integrated versus standard treatment for patients
with a first episode of psychotic illness. BMJ. 2005;331:602.
Phillips LJ, Nelson B, Yuen HP, Francey SM, Simmons M, Stanford C, et al.
Randomized controlled trial of interventions for young people at ultra-high risk of
psychosis: study design and baseline characteristics. Australian and New Zealand
Phutane VH, Tek C, Chwastiak L, Ratliff JC, Ozyuksel B, Woods SW, et al.
Cardiovascular risk in a first-episode psychosis sample: a 'critical period' for
prevention? Schizophrenia Research. 2011;127:257-61.
Pierides M. Mental health services in Cyprus. Psychiatric Bulletin. 1994;18:425-27.
Pilling S, Bebbington P, Kuipers E, Garety P, Geddes J, Martindale B, et al.

Psychological treatments in schizophrenia: II. Meta-analyses of randomized
controlled trials of social skills training and cognitive remediation. Psychological
medicine. 2002;32:783-91.
Pique TW. Cost-effectiveness of an African American Focus Assertive Community

Treatment Program. Alameda: California School of Professional Psychology; 1999.
Pitt L, Kilbride M, Welford M, Nothard S, Morrison AP. Impact of a diagnosis of

psychosis: user-led qualitative study. Psychiatric Bulletin. 2009;33:419-23.
Polak PR, Kirby MW, Deitchman WS. Treating acutely psychotic patients in private
homes. New Directions for Mental Health Services. 1979;1:49-64.
Posner CM, Wilson KG, Kral MJ, Lander S, McIlwraith RD. Family
psychoeducational support groups in schizophrenia. American Journal of
Orthopsychiatry. 1992;62:206-18.
Potter WZ, Ko GN, Zhang LD, Yan WW. Clozapine in China - a review and preview
of US/PRC collaboration. Psychopharmacology. 1989;99:S87-S91.

Power P, Iacoponi E, Reynolds N, Fisher H, Russell M, Garety P, et al. The Lambeth
early onset crisis assessment team study: general practitioner education and access
to an early detection team in first-episode psychosis. The British Journal of
Psychiatry. 2007;51:s133-s39.
Priebe S, Badesconyi A, Fioritti A, Hansson L, Kilian R, Torres-Gonzales F, et al.

Reinstitutionalisation in mental health care: comparison of data on service provision
from six European countries. BMJ. 2005;330:123-6.
Priebe S, Katsakou C, Amos T, Leese M, Morriss R, Rose D, et al. Patients' views and
readmissions 1 year after involuntary hospitalisation. The British Journal of
Psychiatry. 2009;194:49-54.
Prince M, Patel V, Saxena S, Maj M, Maselko J, Phillips MR, et al. No health without
mental health. The Lancet. 2007;370:859-77.
Querido A. Community mental hygiene in the city of Amsterdam. Mental Hygiene.

1935;19:177-95.
Quinlivan R, Hough R, Crowell A, Beach C, Hofstetter R, Kenworthy K. Service

utilization and costs of care for severely mentally ill clients in an intensive case
management program. Psychiatric Services. 1995;46:365-71.
Rathod S, Kingdon D, Smith P, Turkington D. Insight into schizophrenia: the effects

of cognitive behavioural therapy on the components of insight and association with
sociodemographics—data on a previously published randomised controlled trial.
Read J, Agar K, Argyle N, Aderhold V. Sexual and physical abuse during childhood
and adulthood as predictors of hallucinations, delusions and thought disorder.
Psychology and Psychotherapy. 2003;76:1-22.
Read J, Bentall RP. Negative childhood experiences and mental health: theoretical,
clinical and primary prevention implications. British Journal of Psychiatry.
2012;200:89-91.
Read J, Haslam N, Sayce L, Davies E. Prejudice and schizophrenia: a review of the

'mental illness is an illness like any other' approach. Acta Psychiatrica Scandinavica.
2006;114:303-18.
Read J, van Os J, Morrison AP, Ross CA. Childhood trauma, psychosis and
schizophrenia: a literature review with theoretical and clinical implications. Acta
Rector NA, Seeman MV, Segal ZV. Cognitive therapy for schizophrenia: a
preliminary randomized controlled trial. Schizophrenia Research. 2003;63:1-11.

Reid J, Lloyd C, de Groot L. The psychoeducation needs of parents who have an
adult son or daughter with a mental illness. Australian e-Journal for the
Advancement of Mental Health. 2005;4:65-67.
Reilly S, Planner C, Hann M, Reeves D, Nazareth I, Lester H. The role of primary

care in service provision for people with severe mental illness in the United
Kingdom. PLoS One. 2012;7:e36468.
Reinares M, Vieta E, Colom F, Martínez-Arán A, Torrent C, Comes M, et al. Impact

of a psychoeducational family intervention on caregivers of stabilized bipolar
patients. Psychotherapy and Psychosomatics. 2004:312-9.
Remington G, Kapur S, Zipursky RB. Pharmacotherapy of first-episode

schizophrenia. The British Journal of Psychiatry. 1998;172:66-70.
Renwick L, Gavin B, McGlade N, Lacey P, Goggins R, Jackson D, et al. Early

intervention service for psychosis: views from primary care. Early Intervention in
Psychiatry. 2008;2:285-90.
Repper J, Carter T. Using personal experience to support others with similar

difficulties: a review of the literature on peer support in mental health services.
Nottingham: Together and the University of Nottingham. 2010.
Repper J, Perkins R. Social Inclusion and Recovery: A Model for Mental Health
Practice. Edinburgh: Balliere Tindall; 2003.
Repper J, Watson E. A year of peer support in Nottingham: lessons learned. Journal

of Mental Health Training, Education and Practice. 2012;7:70-78.
Revicki DA, Shakespeare A, Kind P. Preferences for schizophrenia-related health

states: a comparison of patients, caregivers and psychiatrists. International Clinical
Rice CD, Howard L, Leese M, Jarrett M, Thornicroft G. Determinants of wanting to

seek full versus part-time paid employment among people with severe mental
illness. Journal of Mental Health 2009;18:424-32.
Riley G, Gregory N, Bellinger J, Davies N, Mabbott G, Sabourin R. Carer's education

groups for relatives with a first episode of psychosis: An evaluation of an eight-week
education group. Early Intervention in Psychiatry. 2011;5:57-63.
Rinaldi M, Miller L, Perkins R. Implementing the individual placement and support

(IPS) approach for people with mental health conditions in England. International
Review of Psychiatry. 2010;22:163-72.

Rivera JJ, Sullivan AM, Valenti SS. Adding consumer-providers to intensive case
management: does it improve outcome? Psychiatric Services. 2007;58:802-09.
Robinson D, Woerner MG, Alvir JM, Bilder R, Goldman R, Geisler S, et al. Predictors
of relapse following response from a first episode of schizophrenia or schizoaffective
disorder. Archives of General Psychiatry. 1999;56:241-7.
Rogers ES, Teague GB, Lichenstein C, Campbell J, Lyass A, Chen R, et al. Effects of
participation in consumer-operated service programs on both personal and
organizationally mediated empowerment: results of multisite study. Journal of
Rehabilitation Research and Development. 2007;44:785-800.
Rohricht F, Priebe S. Effect of body-oriented psychological therapy on negative

symptoms in schizophrenia: a randomized controlled trial. Psychological medicine.
2006;36:669-78.
Roick C, Heider D, Bebbington PE, Angermeyer MC, Azorin J-M, Brugha TS, et al.
Burden on caregivers of people with schizophrenia: comparison between Germany
and Britain. The British Journal of Psychiatry. 2007;190:333-38.
Rooney R, Wright B, O'Neil K. Issues faced by carers of people with a mental illness
from culturally and linguistically diverse backgrounds: carers' and practitioners'
perceptions. Australian e-Journal for the Advancement of Mental Health. 2006;5:1-
13.
Rosen K, Garety P. Predicting recovery from schizophrenia: a retrospective

comparison of characteristics at onset of people with single and multiple episodes.
Rosenfarb IS, Bellack AS, Aziz N. A sociocultural stress, appraisal, and coping model
of subjective burden and family attitudes toward patients with schizophrenia.
Journal of Abnormal Psychology. 2006;115:157-65.
Rosenheck R, Neale M, Gallup P. Community-oriented mental health care: assessing

diversity in clinical practice. Psychosocial Rehabilitation Journal. 1993;16:39-50.
Rosenheck R, Perlick D, Bingham S, Liu-Mares W, Collins J, Warren S, et al.

Effectiveness and cost of olanzapine and haloperidol in the treatment of
schizophrenia: a randomized controlled trial. JAMA: The Journal of the American
Medical Association. 2003;290:2693-702.
Rosenheck RA, Cramer J, Xu W, Thomas J, Henderson W, Frisman L, et al. A

comparison of clozapine and haloperidol in hospitalised patients with refractory
schizophrenia. The New England Journal of Medicine. 1997;337:809-15.

Rosenheck RA, Leslie DL, Sindelar J, Miller EA, Lin H, Stroup TS, et al. Cost-
effectiveness of second-generation antipsychotics and perphenazine in a randomized
trial of treatment for chronic schizophrenia. American Journal of Psychiatry.
2006;163:2080-9.
Rosie JS. Partial hospitalization: a review of recent literature. Hospital and

Ross J. Occupational Therapy and Vocational Rehabilitation. Chichester: John Wiley

& Sons Ltd; 2008.
Rosser R, Cottee M, Rabin R, Selai C. Index of health-related quality of life. In:

Hopkins A, ed. Measures of the Quality of Life and the Uses to Which Such
Measures May Be Put. London: The Royal College of Physicians; 1992.
Rostami-Hodjegan A, Amin AM, Spencer EP, Lennard MS, Tucker GT, Flanagan RJ.
Influence of dose, cigarette smoking, age, sex, and metabolic activity on plasma
clozapine concentrations: a predictive model and nomograms to aid clozapine dose
adjustment and to assess compliance in individual patients. Journal of Clinical
Roth A, Fonagy P, Parry G. What Works for Whom? A Critical Review of

Psychotherapy Research. New York: Guilford; 1996.
Royal College of Psychiatrists. Consensus Statement on High-Dose Antipsychotic

Medication. London: Royal College of Psychiatrists; 2006.
Royal College of Psychiatrists. Improving services for refugees and asylum seekers:
position statement. London: Royal College of Psychiatrists; 2007.
Royal College of Psychiatrists. Personal communication with the Prescribing

Observatory for Mental Health [2005 data]. Regular Oral Daily Prescriptions. 2008.
Royal College of Psychiatrists. Carers and Confidentiality. 2010; Available from:

http://www.rcpsych.ac.uk/about/campaigns/partnersincarecampaign/carersandc
onfidentiality.aspx.
Royal College of Psychiatrists. Report of the National Audit of Schizophrenia (NAS)

2012. London: Healthcare Quality Improvement Partnership; 2012.
Ruhrmann S, Bechdolf A, Kuhn KU, Wagner M, Schultze-Lutter F, Janssen B, et al.

Acute effects of treatment for prodromal symptoms for people putatively in a late
initial prodromal state of psychosis. The British Journal of Psychiatry. 2007;51:s88-95.
Ruhrmann S, Schultze-Lutter F, Salokangas R, Heinimaa M, Linszen D, Dingemans

P, et al. Prediction of psychosis in adolescents and young adults at high risk: results

from the prospective European prediction of psychosis study. Archives of General
Psychiatry. 2010;67:241.
Russo M, Levine SZ, Demjaha A, Di Forti M, Bonaccorso S, Fearon P, et al.

Association between symptom dimensions and categorical diagnoses of psychosis: a
cross-sectional and longitudinal investigation. Schizophrenia Bulletin.
2013;DOI:10.1093/schbul/sbt055. Published online: 9 May.
Saari K, Koponen H, Laitinen J, Jokelainen J, Lauren L, Isohanni M, et al.

Hyperlipidemia in persons using antipsychotic medication: a general population-
based birth cohort study. Journal of Clinical Psychiatry. 2004;65:547-50.
Saari KM, Lindeman SM, Viilo KM, Isohanni MK, Jarvelin MR, Lauren LH, et al. A 4-
fold risk of metabolic syndrome in patients with schizophrenia: the Northern
Finland 1966 Birth Cohort study. Journal of Clinical Psychiatry. 2005;66:559-63.
Sacco KA, Creeden C, Reutenauer EL, Vessicchio JC, Weinberger AH, George TP.
Effects of atomoxetine on cognitive function and cigarette smoking in schizophrenia.
Saddichha S, Manjunatha N, Ameen S, Akhtar S. Diabetes and schizophrenia - effect

of disease or drug? Results from a randomized, double-blind, controlled prospective
study in first-episode schizophrenia. Acta Psychiatrica Scandinavica. 2008;117:342-7.
Saha S, Chant D, McGrath J. A systematic review of mortality in schizophrenia: Is the

differential mortality gap worsening over time? Archives of General Psychiatry.
2007;64:1123-31.
Sainsbury Centre for Mental Health. Paper 3: the economic and social costs of mental
illness. London: Sainsbury Centre for Mental Health; 2003.
Sainsbury Centre for Mental Health. The Economic and Financial Case for
Supported Employment. London: Sainsbury Centre for Mental Health; 2009.
Salkever D, Domino ME, Burns BJ, Santos AB, Deci PA, Dias J, et al. Assertive
community treatment for people with severe mental illness: the effect on hospital use
and costs. Health Services Research. 1999;34:577-601.
Salyers MP, McGuire AB, Rollins AL, Bond GR, Mueser KT, Macy VR. Integrating
assertive community treatment and illness management and recovery for consumers
with severe mental illness. Community Mental Health Journal. 2010;46:319-29.
Salzer MS, Shear SL. Identifying consumer-provider benefits in evaluations of

consumer-delivered services. Psychiatric Rehabilitation Journal. 2002;25:281-8.
Sartorius N. Iatrogenic stigma of mental illness. BMJ. 2002;324:1470-1.

Sass B, Moffat J, Bhui K, Mckenzie K. Enhancing pathways to care for black and
minority ethnic populations: a systematic review. International Review of
Psychiatry. 2009;21:430-38.
Sattelmair JR, Perman J, Ding EL, Kohl III HW, Haskell W, Lee I. Dose response
between physical activity and risk of coronary heart disease: a meta-analysis.
Circulation. 2011;124:789-95.
Saunders JC, Byrne MM. A thematic analysis of families living with schizophrenia.
Archives of Psychiatric Nursing. 2002;16:217-23.
Scheewe TW, Backx FJG, Takken T, Jorg F, van Strater ACP, Kroes AG, et al. Exercise
therapy improves mental and physical health in schizophrenia: A randomised
controlled trial. Acta Psychiatrica Scandinavica. 2013;127:464-73.
Schizophrenia Commission. The abandoned illness: a report from the Schizophrenia

Commission. London: Rethink Mental Illness; 2012.
Schneider J, Boyce M, Johnson R, Secker J, Slade J, Grove B, et al. Impact of

supported employment on service costs and income of people with mental health
needs Journal of Mental Health. 2009;18:533-42.
Schneider J, Secker J, Grove B., Floyd M, Boyce M, Practice Partners. The SESAMI
evaluation of employment support in the UK: background and baseline data. Journal
of Mental Health 2007;16:375-87.
Schooler NR. Relapse and rehospitalization: comparing oral and depot

antipsychotics. Journal of Clinical Psychiatry. 2003;64(Suppl:16):14-7.
Schünemann H, Brożek J, Oxman A, eds. GRADE Handbook: for Grading the

Quality of Evidence and Strength of Recommendations. Version 3.2: The GRADE
Working Group; 2009. Available from:
www.who.int/hiv/topics/mtct/grade_handbook.pdf.
Schünemann HJ, Best D, Vist G, for the GRADE Working Group. Letters, numbers,
symbols and words: how to communicate grades of evidence and recommendations.
Canadian Medical Association Journal. 2003;169:677–80.
Sciolla A, Patterson TL, Wetherell JL, McAdams LA, Jeste DV. Functioning and well-
being of middle-aged and older patients with schizophrenia: measurement with the
36-item Short-Form (SF-36) Health Survey. American Journal of Geriatric Psychiatry.
2003;11:629-37.
Scocco P, Longo R, Caon F. Weight change in treatment with olanzapine and a

psychoeducational approach. Eating Behaviors. 2006;7:115-24.

Scottish Recovery Network. The role and potential development of peer support
services: Scottish Recovery Network briefing paper. Glasgow: Scottish Recovery
Network; 2005.
Seebohm P, Secker J. What do service users want? In: Grove B, Secker J, Seebohm P,
eds. New Thinking about Mental Health and Employment. Oxford: Radcliffe
Publishing; 2005.
Segal SP, Silverman C, Temkin T. Outcomes from consumer-operated and

community mental health services: A randomized controlled trial. Psychiatric
Services. 2011;62:915-21.
Sells D, Davidson L, Jewell C, Falzer P, Rowe M. The treatment relationship in peer-

based and regular case management for clients with severe mental illness.
Serretti A, Mandelli L, Bajo E, Cevenini N, Papili P, Mori E, et al. The socio-

economical burden of schizophrenia: a simulation of cost-offset of early intervention
program in Italy. European Psychiatry. 2009;24:11-6.
Sevy S, Nathanson K, Schechter C, Fulop G. Contingency valuation and preferences

of health states associated with side effects of antipsychotic medications in
Shalev A, Hermesh H, Rothberg J, Munitz H. Poor neuroleptic response in acutely

exacerbated schizophrenic patients. Acta Psychiatrica Scandinavica. 1993;87:86-91.
Sharif F, Shaygan M, Mani A. Effect of a psycho-educational intervention for family

members on caregiver burdens and psychiatric symptoms in patients with
schizophrenia in Shiraz, Iran. BMC Psychiatry. 2012;12.
Shepherd G. Social skills training: the generalisation problem-some further data.

Behaviour Research and Therapy. 1978;16:297-9.
Shern DL, Tsemberis S, Anthony W, Lovell AM, Richmond L, Felton CJ, et al.
Serving street-dwelling individuals with psychiatric disabilities: outcomes of a
psychiatric rehabilitation clinical trial. American Journal of Public Health.
2000;90:1873-78.
Shevlin M, Houston JE, Dorahy MJ, Adamson G. Cumulative traumas and

psychosis: an analysis of the national comorbidity survey and the British psychiatric
morbidity survey. Schizophrenia Bulletin. 2008;34:193-9.
Shiloh R, Zemishlany Z, Aizenberg D, Radwan M, Schwartz B, Dorfman-Etrog P, et

al. Sulpiride augmentation in people with schizophrenia partially responsive to

clozapine. A double-blind, placebo-controlled study. The British Journal of
Psychiatry. 1997;171:569-73.
Shon KH, Park SS. Medication and symptom management education program for
the rehabilitation of psychiatric patients in Korea: the effects of promoting schedule
on self-efficacy theory. Yonsei Medical Journal. 2002;43:579-89.
Simon AE, Lauber C, Ludewig K, Braun-Scharm H, Umbricht DS, Swiss Early

Psychosis Project. General practitioners and schizophrenia: results from a Swiss
survey. The British Journal of Psychiatry. 2005;187:274-81.
Simpson A, Janner M. Star Wards Survey Report 2009/2010. London: Star Wards;
2010. Available from:
http://www.starwards.org.uk/images/stories/user_files/marion/Star_Wards_Sur
vey_Report__Final_05052010_1.pdf.
Singh SP, Fisher HL. Early intervention in psychosis: obstacles and opportunities.
Advances in Psychiatric Treatment. 2005;11:71-78.
Singleton N, Bumpstead R, O'Brien M, Lee A, Meltzer H. Psychiatric morbidity

among adults living in private households, 2000. International Review of Psychiatry.
2003;15:65-73.
Skovholt TM. The client as helper: a means to promote psychological growth. The
Counseling Psychologist. 1974;4:58-64.
Skrinar GS, Huxley NA, Hutchinson DS, Menninger E, Glew P. The role of a fitness
intervention on people with serious psychiatric disabilities. Psychiatric
Rehabilitation Journal. 2005;29:122-27.
Slade E, McCarthy J, Valenstein M, Visnic S, Dixon L. Cost savings from assertive

community treatment services in an era of declining psychiatric inpatient use. Health
Services Research. 2013;48:195-217.
Slade M. 100 Ways to Support Recovery: a Guide for Mental Health Professionals.
London: Rethink; 2009.
Slade M, Rosen A, Shankar R. Multidisciplinary mental health teams. The

International Journal of Social Psychiatry. 1995;41:180-9.
Slade PD, Bentall RP. Sensory Deception: A Scientific Analysis of Hallucination.

London: Croom Helm; 1988.
Sledge WH, Lawless M, Sells D, Wieland M, O'Connell MJ, Davidson L.

Effectiveness of peer support in reducing readmissions of persons with multiple
psychiatric hospitalizations. Psychiatric Services. 2011;62:541-4.

Small N, Harrison J, Newell R. Carer burden in schizophrenia: considerations for
nursing practice. Mental Health Practice. 2010;14:22-25.
Smith DJ, Langan J, McLean G, Guthrie B, Mercer SW. Schizophrenia is associated

with excess multiple physical-health comorbidities but low levels of recorded
cardiovascular disease in primary care: cross-sectional study. BMJ. 2013;3.
Smith JV, Birchwood MJ. Specific and non-specific effects of educational intervention
with families living with a schizophrenic relative. The British Journal of Psychiatry.
1987:645-52.
Smith M, Hopkins D, Peveler RC, Holt RI, Woodward M, Ismail K. First- v. second-
generation antipsychotics and risk for diabetes in schizophrenia: systematic review
and meta-analysis. The British Journal of Psychiatry. 2008;192:406-11.
Snyder SH, Greenberg D, Yamumura HI. Antischizophrenic drugs: affinity for

muscarinic cholinergic receptor sites in the brain predicts extrapyramidal effects.
Journal of Psychiatric Research. 1974;11:91-5.
So HW, Chen EYH, Chan RCK, Wong CW, Hung SF, Chung DWS, et al. Efficacy of a
brief intervention for carers of people with first-episode psychosis: A waiting list
controlled study. Hong Kong Journal of Psychiatry. 2006;16:92-100.
Solomon P. Peer support/peer provided services underlying processes, benefits, and

critical ingredients. Psychiatric Rehabilitation Journal. 2004;27:392-401.
Solomon P, Draine J. One-year outcomes of a randomized trial of consumer case

management. Evaluation and Program Planning. 1995;18:117-27.
Solomon P, Draine J, Mannion E, Meisel M. Impact of brief family psychoeducation

on self-efficacy. Schizophrenia Bulletin. 1996;22:41-50.
Solomon P, Draine J, Meyerson A. Jail recidivism and receipt of community mental

health services. Hospital and Community Psychiatry. 1994;45:793-97.
Spiegelhalter D, Thomas A, Best N. WinBUGS Beta Version 1.4 User Manual.

Cambridge: University of Cambridge: MRC Biostatistics Unit, Institute of Public
Health; 2001.
Squires H, Rick J, Carroll C, Hillage J. Cost-effectiveness of interventions to return

employees to work following long-term sickness absence due to musculoskeletal
disorders. Journal of Public Health. 2012;34:115-24.
Stack-Sullivan H. Schizophrenia as a Human Process. London: Norton; 1974.

Stack Sullivan H. Conceptions of Modern Psychiatry. Washington, DC: William
Alanson White Psychiatric Foundation; 1947.
Stafford MR, Jackson H, Mayo-Wilson E, Morrison AP, Kendall T. Early

interventions to prevent psychosis: systematic review and meta-analysis. BMJ.
2013;346:f185.
Stahl SM. Focus on antipsychotic polypharmacy: evidence-based prescribing or

prescribing-based evidence? The International Journal of
Neuropsychopharmacology. 2004;7:113-6.
Startup M, Jackson MC, Bendix S. North Wales randomized controlled trial of

cognitive behaviour therapy for acute schizophrenia spectrum disorders: outcomes
at 6 and 12 months. Psychological medicine. 2004;34:413-22.
Startup M, Jackson MC, Evans KE, Bendix S. North Wales randomized controlled
trial of cognitive behaviour therapy for acute schizophrenia spectrum disorders:
two-year follow-up and economic evaluation. Psychological medicine. 2005;35:1307-
16.
Steel C. The relationship between trauma and psychosis: a CBT perspective.

Reading: Charlie Waller Institute of Evidence Based Psychological Treatment:
University of Reading; 2011. Available from:
http://www.ukpts.co.uk/site/assets/Steel-UKPTS-Oxford-2011.pdf.
Stein LI, Test MA. Alternative to mental hospital treatment. I. Conceptual model,
treatment program, and clinical evaluation. Archives of General Psychiatry.
1980;37:392-7.
Stein LI, Test MA, Marx AJ. Alternative to the hospital: a controlled study. American
Steingard S, Allen M, Schooler NR. A study of the pharmacologic treatment of

medication-compliant schizophrenics who relapse. Journal of Clinical Psychiatry.
1994;55:470-2.
Strakowski SM, Johnson JL, Delbello MP, Hamer RM, Green AI, Tohen M, et al.
Quality of life during treatment with haloperidol or olanzapine in the year following
a first psychotic episode. Schizophrenia Research. 2005;78:161-9.
Stratton IM, Adler AI, Neil HA, Matthews DR, Manley SE, Cull CA, et al.
Association of glycaemia with macrovascular and microvascular complications of
type 2 diabetes (UKPDS 35): prospective observational study. BMJ. 2000;321:405-12.
Stroup TS, McEvoy JP, Swartz MS, Byerly MJ, Glick ID, Canive JM, et al. The
national institute of mental health clinical antipsychotic trials of intervention

effectiveness (CATIE) project: schizophrenia trial design and protocol development.
Stuckler D, Basu S, Suhrcke M, Coutts A, McKee M. Effects of the 2008 recession on

health: a first look at European data. The Lancet. 2011;378:124-5.
Sullivan PF, Kendler KS, Neale MC. Schizophrenia as a complex trait: evidence from
a meta-analysis of twin studies. Archives of General Psychiatry. 2003;60:1187-92.
Sun SX, Liu GG, Christensen DB, Fu AZ. Review and analysis of hospitalization
costs associated with antipsychotic nonadherence in the treatment of schizophrenia
in the United States. Current Medical Research and Opinion. 2007;23:2305-12.
Suvisaari JM, Saarni SI, Perala J, Suvisaari JV, Harkanen T, Lonnqvist J, et al.
Metabolic syndrome among persons with schizophrenia and other psychotic
disorders in a general population survey. Journal of Clinical Psychiatry.
2007;68:1045-55.
Sytema S, Wunderink L, Bloemers W, Roorda L, Wiersma D. Assertive community

treatment in the Netherlands: a randomized controlled trial. Acta Psychiatrica
Szmukler G, Kuipers E, Joyce J, Harris T, Leese M, Maphosa W, et al. An exploratory

randomised controlled trial of a support programme for carers of patients with a
psychosis. Social Psychiatry and Psychiatric Epidemiology. 2003;38:411-18.
Szmukler GI, Herrman H, Colusa S, Benson A, Bloch S. A controlled trial of a

counselling intervention for caregivers of relatives with schizophrenia. Social
Talwar N, Crawford MJ, Maratos A, Nur U, McDermott O, Procter S. Music therapy

for in-patients with schizophrenia: exploratory randomised controlled trial. The
Tandon R, Keshavan MS, Nasrallah HA. Schizophrenia, "just the facts" what we
know in 2008. 2. Epidemiology and etiology. Schizophrenia Research. 2008;102:1-18.
Tang YL, Mao P, Li FM, Li W, Chen Q, Jiang F, et al. Gender, age, smoking
behaviour and plasma clozapine concentrations in 193 Chinese inpatients with
schizophrenia. British Journal of Clinical Pharmacology. 2007;64:49-56.
Tanskanen S, Morant N, Hinton M, Lloyd-Evans B, Crosby M, Killaspy H, et al.

Service user and carer experiences of seeking help for a first episode of psychosis: a
UK qualitative study. BMC Psychiatry. 2011;11:157.

Tarricone I, Ferrari-Gozzi B, Serretti A, Grieco D, Berardi D. Weight gain in
antipsychotic-naive patients: a review and meta-analysis. Psychological medicine.
2010;40:187-200.
Tarrier N, Beckett R, Harwood S, Baker A, Yusupoff L, Ugarteburu I. A trial of two

cognitive-behavioural methods of treating drug-resistant residual psychotic
symptoms in schizophrenic patients: I. Outcome. The British Journal of Psychiatry.
1993;162:524-32.
Tarrier N, Lowson K, Barrowclough C. Some aspects of family interventions in

schizophrenia. II: financial considerations. The British Journal of Psychiatry.
1991;159:481-84.
Tauscher J, Kapur S. Choosing the right dose of antipsychotics in schizophrenia:

lessons from neuroimaging studies. CNS Drugs. 2001;15:671-8.
Taylor D, Mace S, Mir S, Kerwin R. A prescription survey of the use of atypical

antipsychotics for hospital inpatients in the United Kingdom. International Journal
of Psychiatry in Clinical Practice. 2000;4:41-46.
Taylor D, Mir S, Mace S, Whiskey E. Co-prescribing of atypical and typical

antipsychotics — prescribing sequence and documented outcome. Psychiatric
Bulletin. 2002;26:170-72.
Taylor D, Young C, Mohamed R, Paton C, Walwyn R. Undiagnosed impaired fasting

glucose and diabetes mellitus amongst inpatients receiving antipsychotic drugs.
Journal of Psychopharmacology. 2005;19:182-6.
Taylor MF. Brice J, Buck N, Prentice-Lane E. British Household Panel Survey User
Manual Volume B. Colchester: University of Essex; 2003.
Taylor PJ, Gunn J. Homicides by people with mental illness: myth and reality. The
Test MA, Knoedler WH, Allness DJ, Burke SS. Long-term community care through
an assertive continuous treatment team. In: Tamminga C, Schultz, S., ed. Advances
in Neuropsychiatry and Psychopharmacology, Schizophrenia Research. NY: Raven
Press; 1991.
Thakore JH. Metabolic syndrome and schizophrenia. The British Journal of

Psychiatry. 2005;186:455-6.
The Cochrane Collaboration. Review Manager (RevMan) [Computer programme].

Version 5.1. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration;
2011.

The Health and Social Care Information Centre. Hospital Episode Statistics 2011–12.
London: The Health and Social Care Information Centre; 2012; Available from:
http://www.hesonline.nhs.uk
The International Schizophrenia Consortium. Rare chromosomal deletions and

duplications increase risk of schizophrenia. Nature. 2008;455:237-41.
The NHS Information Centre, Lifestyles Statistics. Statistics on smoking: England.

The Health and Social Care Information Centre; 2011; 1-126:[Available from:
https://catalogue.ic.nhs.uk/publications/public-health/smoking/smok-eng-
2011/smok-eng-2011-rep.pdf.
The Royal College of Psychiatrists Social Inclusion Scoping Group. Mental Health
and Social Inclusion: making psychiatry and mental health services fit for the 21st
century. London: Royal College of Psychiatrists 2009.
The Work Foundation. Working with Schizophrenia: Pathways to Employment,

Recovery and Inclusion. London: The Work Foundation; 2013.
Thieda P, Beard S, Richter A, Kane J. An economic review of compliance with

medication therapy in the treatment of schizophrenia. Psychiatric Services.
2003;54:508-16.
Thomas J, Harden A. Methods for the thematic synthesis of qualitative research in

systematic reviews. BMC Medical Research Methodology. 2008;8:45.
Thorne B. Carl Rogers. London: Sage; 1992.
Thornicroft G. Shunned: discrimination against people with mental illness. Oxford

and New York: Oxford University Press; 2006.
Thornicroft G, Tansella M. The balanced care model for global mental health.
Thornicroft G, Tansella M, Becker T, Knapp M, Leese M, Schene A, et al. The

personal impact of schizophrenia in Europe. Schizophrenia Research. 2004;69:125-32.
Tiihonen J, Wahlbeck K, Lonnqvist J, Klaukka T, Ioannidis JP, Volavka J, et al.

Effectiveness of antipsychotic treatments in a nationwide cohort of patients in
community care after first hospitalisation due to schizophrenia and schizoaffective
disorder: observational follow-up study. BMJ. 2006;333:224.
Tilden D, Aristides M, Meddis D, Burns T. An economic assessment of quetiapine

and haloperidol in patients with schizophrenia only partially responsive to
conventional antipsychotics. Clinincal Therapeutics. 2002;24:1648-67.

Torrey EF, Bartko JJ, Yolken RH. Toxoplasma gondii and other risk factors for
schizophrenia: an update. Schizophrenia Bulletin. 2012;38:642-7.
Totman J, Mann F, Johnson S. Is locating acute wards in the general hospital an

essential element in psychiatric reform? The UK experience. Epidemiologia e
Psichiatria Sociale. 2010;19:282-6.
Tranvag O, Kristoffersen K. Experience of being the spouse/cohabitant of a person

with bipolar affective disorder: a cumulative process over time. Scandinavian
Journal of Caring Sciences. 2008;22:5-18.
Trower P, Birchwood M, Meaden A, Byrne S, Nelson A, Ross K. Cognitive therapy

for command hallucinations: randomised controlled trial. The British Journal of
Psychiatry. 2004;184:312-20.
Tsang HW, Chan A, Wong A, Liberman RP. Vocational outcomes of an integrated

supported employment program for individuals with persistent and severe mental
illness. Journal of Behavior Therapy and Experimental Psychiatry. 2009;40:292-305.
Tsoi DT, Porwal M, Webster AC. Interventions for smoking cessation and reduction
in individuals with schizophrenia. Cochrane Database of Systematic Reviews. 2013;2.
Tunis SL, Croghan TW, Heilman DK, Johnstone BM, Obenchain RL. Reliability,
validity, and application of the medical outcomes study 36-item short-form health
survey (SF-36) in schizophrenic patients treated with olanzapine versus haloperidol.
Medical Care. 1999;37:678-91.
Tunis SL, Faries DE, Nyhuis AW, Kinon BJ, Ascher-Svanum H, Aquila R. Cost-
effectiveness of olanzapine as first-line treatment for schizophrenia: results from a
randomized, open-label, 1-year trial. Value in Health. 2006;9:77-89.
Tuomilehto J, Schwarz P, Lindstrom J. Long-term benefits from lifestyle

interventions for type 2 diabetes prevention: time to expand the efforts. Diabetes
Care. 2011;34 (Suppl):S210-S14.
Turkington D, Kingdon D, Turner T. Effectiveness of a brief cognitive-behavioural

therapy intervention in the treatment of schizophrenia. The British Journal of
Psychiatry. 2002;180:523-7.
Turner AP, Barlow JH, Elliot CH. Lay-Led, Self-Management for People with a
Long-Term Health Condition: UK Results. T. Kroll edn. Disability and Health,
Hauppage, New York: Nova Science Publishers, Inc.; 2008.
Turner N, Browne S, Clarke M, Gervin M, Larkin C, Waddington JL, et al.

Employment status amongst those with psychosis at first presentation. Social

Twamley EW, Vella L, Burton CZ, Becker DR, Bell MD, Jeste DV. The efficacy of
supported employment for middle-aged and older people with schizophrenia.
Tyrer P, Evans K, Gandhi N, Lamont A, Harrison RP, Johnson T. Randomised

controlled trial of two models of care for discharged psychiatric patients. BMJ.
1998;316:106-09.
Udechuku A, Olver J, Hallam K, Blyth F, Leslie M, Nasso M, et al. Assertive

community treatment of the mentally ill: service model and effectiveness.
Australasian Psychiatry. 2005;13:129-34.
Ulrich G, Houtmans T, Gold C. The additional therapeutic effect of group music

therapy for schizophrenic patients: a randomized study. Acta Psychiatrica
Upthegrove R, Birchwood M, Ross K, Brunett K, McCollum R, Jones L. The

evolution of depression and suicidality in first episode psychosis. Acta Psychiatrica
Usher K, Park T, Foster K, Buettner P. A randomized controlled trial undertaken to

test a nurse-led weight management and exercise intervention designed for people
with serious mental illness who take second generation antipsychotics. Journal of
Advanced Nursing. 2013;69:1539-48.
Valmaggia LR, McCrone P, Knapp M, Woolley JB, Broome MR, Tabraham P, et al.
Economic impact of early intervention in people at high risk of psychosis.
van Dam DS, van der Ven E, Velthorst E, Selten JP, Morgan C, de Haan L. Childhood
bullying and the association with psychosis in non-clinical and clinical samples: a
review and meta-analysis. Psychological medicine. 2012;42:2463-74.
van den Akker M, Buntinx F, Metsemakers JF, Roos S, Knottnerus JA.

Multimorbidity in general practice: prevalence, incidence, and determinants of co-
occurring chronic and recurrent diseases. Journal of Clinical Epidemiology.
1998;51:367-75.
Van Gent EM, Zwart FM. Psychoeducation of partners of bipolar-manic patients.

Journal of Affective Disorders. 1991;21:15-8.
Van Gestel-Timmermans H, Brouwers EPM, Van Assen MALM, Van

Nieuwenhuizen C. Effects of a peer-run course on recovery from serious mental
illness: a randomized controlled trial. Psychiatric Services. 2012;63:54-60.

Van Nimwegen LJ, Storosum JG, Blumer RM. Hepatic insulin resistance in
antipsychotic naive schizophrenic patients: stable isotope studies of glucose
metabolism. The Journal of Clinical Endocrinology and Metabolism. 2008:572–77.
van Os J, Kenis G, Rutten BP. The environment and schizophrenia. Nature.

2010;468:203-12.
van Os J, Linscott RJ, Myin-Germeys I, Delespaul P, Krabbendam L. A systematic

review and meta-analysis of the psychosis continuum: evidence for a psychosis
proneness-persistence-impairment model of psychotic disorder. Psychological
medicine. 2009;39:179-95.
van Winkel R, De Hert M, Van Eyck D, Hanssens L, Wampers M, Scheen A, et al.

Screening for diabetes and other metabolic abnormalities in patients with
schizophrenia and schizoaffective disorder: evaluation of incidence and screening
methods. Journal of Clinical Psychiatry. 2006;67:1493-500.
van Winkel R, De Hert M, Wampers M, Van Eyck D, Hanssens L, Scheen A, et al.

Major changes in glucose metabolism, including new-onset diabetes, within 3
months after initiation of or switch to atypical antipsychotic medication in patients
with schizophrenia and schizoaffective disorder. Journal of Clinical Psychiatry.
2008;69:472-79.
Vancampfort D, Probst M, Sweers K, Maurissen K, Knapen J, De Hert M.

Relationships between obesity, functional exercise capacity, physical activity
participation and physical self-perception in people with schizophrenia. Acta
Varambally S, Gangadhar BN, Thirthalli J, Jagannathan A, Kumar S,

Venkatasubramanian G, et al. Therapeutic efficacy of add on yogasana intervention
in stabilized outpatient schizophrenia: randomized controlled comparison with
exercise and waitlist. Indian Journal of Psychiatry. 2012;54:227-32.
Varese F, Smeets F, Drukker M, Lieverse R, Lataster T, Viechtbauer W, et al.

Childhood adversities increase the risk of psychosis: a meta-analysis of patient-
control, prospective- and cross-sectional cohort studies. Schizophrenia Bulletin.
2012;38:661-71.
Varghese D, Scott J, Welham J, Bor W, Najman J, O'Callaghan M, et al. Psychotic-like

experiences in major depression and anxiety disorders: a population-based survey in
young adults. Schizophrenia Bulletin. 2011;37:389-93.
Vassos E, Pedersen CB, Murray RM, Collier DA, Lewis CM. Meta-analysis of the
association of urbanicity with schizophrenia. Schizophrenia Bulletin. 2012;38:1118-
23.

Vaughn CE, Leff JP. The influence of family and social factors on the course of
psychiatric illness. A comparison of schizophrenic and depressed neurotic patients.
Vauth R, Corrigan PW, Clauss M, Dietl M, Dreher RM, Stieglitz RD, et al. Cognitive
strategies versus self-management skills as adjunct to vocational rehabilitation.
Vera-Llonch M, Delea TE, Richardson E, Rupnow M, Grogg A, Oster G. Outcomes

and costs of risperidone versus olanzapine in patients with chronic schizophrenia or
schizoaffective disorders: a Markov model. Value in Health. 2004;7:569-84.
Viguera AC, Baldessarini RJ, Hegarty JD, van Kammen DP, Tohen M. Clinical risk
following abrupt and gradual withdrawal of maintenance neuroleptic treatment.
von Hausswolff-Juhlin Y, Bjartveit M, Lindstrom E, Jones P. Schizophrenia and

physical health problems. Acta Psychiatrica Scandinavica Supplementum. 2009:15-
21.
Vreeland B, Minsky S, Yanos PT, Menza M, Gara M, Kim E, et al. Efficacy of the
team solutions program for educating patients about illness management and
treatment. Psychiatric Services. 2006;57:822-28.
Wahlbeck K, Cheine MV, Essali A. Clozapine versus typical neuroleptic medication

for schizophrenia (Cochrane Review). The Cochrane Library, Issue 3. Oxford:
Update Software; 1999.
Wahlbeck K, Westman J, Nordentoft M, Gissler M, Laursen TM. Outcomes of Nordic

mental health systems: life expectancy of patients with mental disorders. The British
Wahlberg KE, Wynne LC, Oja H, Keskitalo P, Pykalainen L, Lahti I, et al. Gene-
environment interaction in vulnerability to schizophrenia: findings from the Finnish
adoptive family study of schizophrenia. American Journal of Psychiatry.
1997;154:355-62.
Wainwright L, Glentworth D, Haddock G, Bentley R, Lobban F. What do relatives

experience when supporting someone in early psychosis. In press.
Walburn J, Gray R, Gournay K, Quraishi S, David AS. Systematic review of patient

and nurse attitudes to depot antipsychotic medication. The British Journal of
Psychiatry. 2001;179:300-7.
Walker J, Craissati J, Batson S, Amos T, Knowles P. How to get better value from
psychiatric units. Health Service Journal 2012.

Walker R, Winick W, Frost ES, Lieberman JM. Social restoration of hospitalized
psychiatric patients through a program of special employment in industry.
Rehabilitation Literature. 1969;30:297-303.
Wallace CJ, Nelson CJ, Liberman RP, Aitchison RA, Lukoff D, Elder JP, et al. A
review and critique of social skills training with schizophrenic-patients.
Wardle M, Drage L, Spencer H, Brabban A, Turkington D, Morrison A. Internalised

stigma, emotional dysfunction and psychotic experiences in people with psychosis.
In press.
Warner R. Recovery from Schizophrenia 2nd edn. NY: Routledge; 1994.
Warner R. The Prevention of Schizophrenia: what interventions are safe and

effective? Schizophrenia Bulletin. 2001;27:551-62.
Weimand BM, Hedelin B, Hall-Lord M-L, Sallstrom C. "Left alone with straining but
inescapable responsibilities": Relatives' experiences with mental health services.
Issues in Mental Health Nursing. 2011;32:703-10.
Weinberger AH, George TP, Perkins KA, Chengappa KN. Effects of topiramate on
smoking in patients with schizoaffective disorder, bipolar type. Journal of Clinical
Weinberger DR, Berman KF, Illowsky BP. Physiological dysfunction of dorsolateral

prefrontal cortex in schizophrenia: III. A new cohort and evidence for a
monoaminergic mechanism. Archives of General Psychiatry. 1988;45:609.
Weiner E, Ball MP, Buchholz AS, Gold JM, Evins AE, McMahon RP, et al. Bupropion
sustained release added to group support for smoking cessation in schizophrenia: a
new randomized trial and a meta-analysis. Journal of Clinical Psychiatry. 2012;73:95-
102.
Weiner E, Buchholz A, Coffay A, Liu F, McMahon RP, Buchanan RW, et al.

Varenicline for smoking cessation in people with schizophrenia: a double blind
randomized pilot study. Schizophrenia Research. 2011;129:94-95.
Weinmann S, Read J, Aderhold V. Influence of antipsychotics on mortality in

schizophrenia: systematic review. Schizophrenia research. 2009;113:1-11.
Welham J, Isohanni M, Jones P, McGrath J. The antecedents of schizophrenia: a

review of birth cohort studies. Schizophrenia Bulletin. 2009;35:603-23.

Wexler BE, Bell MD. Cognitive remediation and vocational rehabilitation for
Wiersma D, Kluiter H, Nienhuis F, Ruphan M, Giel R. Day-treatment with

community care as an alternative to standard hospitalisation: an experiment in the
Netherlands. A preliminary communication. Groningen: Department of Social
Psychiatry, University of Groningen. 1989.
Wiersma D, Nienhuis FJ, Sloof CJ, Giel R. Natural course of schizophrenic disorders:
a 15-year follow-up of a Dutch incidence cohort Schizophrenia Bulletin. 1998;24:75-
85.
Wiersma D, Wanderling J, Dragomirecka E, Ganev K, Harrison G, An Der Heiden

W, et al. Social disability in schizophrenia: its development and prediction over 15
years in incidence cohorts in six European centres. Psychological medicine.
2000;30:1155-67.
Wigman JT, van Nierop M, Vollebergh WA, Lieb R, Beesdo-Baum K, Wittchen HU,
et al. Evidence that psychotic symptoms are prevalent in disorders of anxiety and
depression, impacting on illness onset, risk, and severity-implications for diagnosis
and ultra-high risk research. Schizophrenia Bulletin. 2012;38:247-57.
Williams JM, Anthenelli RM, Morris CD, Treadow J, Thompson JR, Yunis C, et al. A
randomized, doubleblind, placebo-controlled study evaluating the safety and
efficacy of varenicline for smoking cessation in patients with schizophrenia or
schizoaffective disorder. Journal of Clinical Psychiatry and Clinical Neurosciences.
2012a;73:654-60.
Williams JM, Anthenelli RM, Morris CD, Treadow J, Thompson JR, Yunis C, et al. A
randomized, double-blind, placebo-controlled study evaluating the safety and
efficacy of varenicline for smoking cessation in patients with schizophrenia or
schizoaffective disorder. Journal of Clinical Psychiatry. 2012b;73:654-60.
Williams JM, Gandhi KK, Foulds J, Steinberg M, Lou S, Masumova F. No advantage

for high dose compared to regular dose nicotine patch on short-term abstinence rates
in schizophrenia (PA2-3). Society for Research on Nicotine and Tobacco 13th Annual
Meeting:2007; Austin, Texas.
Winterbourne S, Knapp M, McCrone P, Bell N, Campion J, Clark M, et al. Preventing

future physical morbidity and premature mortality in people with first-episode
psychosis: an economic evaluation of the possible benefits of weight management
interventions. 2013a;In press.
Winterbourne S, Knapp M, McCrone P, Bell N, Campion J, Clark M, et al. Quitting

smoking for young people with schizophrenia – is it worth it? Economic evaluation
of smoking cessation interventions. 2013b;In press.

Wirshing DA, Pierre JM, Wirshing WC, Guzik LH, Resnick SA, Goldstein D, et al.
Community re-entry program training module for schizophrenic inpatients
improves treatment outcomes. Schizophrenia Research. 2006;87:338-9.
Witheridge TF. The "active ingredients" of assertive outreach. New Directions for
Mental Health Services. 1991;52:47-64.
Witheridge TF, Dincin J, Appleby L. Working with the most frequent recidivists: a
total team approach to assertive resource management. Psychosocial Rehabilitation
Journal. 1982;5:9–11.
Wolkon GH, Karmen M, Tanaka HT. Evaluation of a social rehabilitation program

for recently released psychiatric patients. Community Mental Health Journal.
1971;7:312-22.
Wong K, Chiu R, Tang B, Mak D, Liu J, Chiu SN. A randomized controlled trial of a
supported employment program for persons with long-term mental illness in Hong
Kong. Psychiatric Services. 2008;59:84-90.
Wood C. The history of art therapy 1938–95. In Art, Psychotherapy and Psychosis. K.
Killick & S. Schaverien edn. London: Routledge; 1997.
World Health Organization. The ICD-10 Classification of Mental and Behavioural

Disorders: Clinical Description and Diagnostic Guidelines. Geneva: World Health
Organization; 1992.
Worthington A, Rooney P, Hannan R. The Triangle of Care. Second Edition. London:

Carers Trust; 2013.
Wright C, Burns T, James P, Billings J, Johnson S, Muijen M, et al. Assertive outreach

teams in London: models of operation. Pan-London Assertive Outreach Study, part
1. The British Journal of Psychiatry. 2003;183:132-8.
Wu EQ, Birnbaum HG, Shi L, Ball DE, Kessler RC, Moulis M, et al. The economic
burden of schizophrenia in the United States in 2002. Journal of Clinical Psychiatry.
2005;66:1122-9.
Wu MK, Wang CK, Bai YM, Huang CY, Lee SD. Outcomes of obese, clozapine-
treated inpatients with schizophrenia placed on a six-month diet and physical
activity program. Psychiatric Services. 2007;58:544-50.
Wu RR, Zhao JP, Jin H, Shao P, Fang MS, Guo XF, et al. Lifestyle intervention and
metformin for treatment of antipsychotic-induced weight gain: a randomized
controlled trial. JAMA : the journal of the American Medical Association.
2008;299:185-93.

Wunderink L, Nieboer RM, Wiersma D, Sytema S, Nienhuis FJ. Recovery in remitted
first-episode psychosis at 7 years of follow-up of an early dose
reduction/discontinuation or maintenance treatment strategy: Long-term follow-up
of a 2-year randomized clinical trial. JAMA Psychiatry. 2013;onlinefirst.
Wykes T, Reeder C. Cognitive Remediation Therapy for Schizophrenia: Theory and

Practice. London: Routledge; 2005.
Wykes T, van der Gaag M. Is it time to develop a new cognitive therapy for
psychosis—cognitive remediation therapy (CRT)? Clinical psychology review.
2001;21:1227-56.
Xiang Y, Weng Y, Li W, Gao L, Chen G, Xie L, et al. Training patients with

schizophrenia with the community re-entry module: a controlled study. Social
Xiang YT, Weng YZ, Li WY, Gao L, Chen GL, Xie L, et al. Efficacy of the Community
Re-Entry Module for patients with schizophrenia in Beijing, China: outcome at 2-
year follow-up. The British Journal of Psychiatry. 2007;190:49-56.
Yagcioglu AEA, Akdede BBK, Turgut TI, Tumuklu M, Yazici MK, Alptekin K, et al.
A double-blind controlled study of adjunctive treatment with risperidone in
schizophrenic patients partially responsive to clozapine: Efficacy and safety. Journal
of Clinical Psychiatry. 2005;66:63-72.
Yang LH, Wonpat-Borja AJ, Opler MG, Corcoran CM. Potential stigma associated
with inclusion of the psychosis risk syndrome in the DSM-V: an empirical question.
Yang W, Li Z, Weng Y, Zhang H, Ma B, Yang W. Psychosocial rehabilitation effects

of music therapy in chronic schizophrenia. Hong Kong Journal of Psychiatry.
1998;8:38-40.
Yoshii H, Watanabe Y, Kitamura H, Nan Z, Akazawa K. Effect of an education

program on improving help-seeking among parents of junior and senior high school
students in Japan. Global Journal of Health Science. 2011;4:33.
Young JL, Spitz RT, Hillbrand M, Daneri G. Medication adherence failure in

schizophrenia: a forensic review of rates, reasons, treatments, and prospects. J Am
Acad Psychiatry Law. 1999;27:426-44.
Young JL, Zonana HV, Shepler L. Medication noncompliance in schizophrenia:

codification and update. Bull Am Acad Psychiatry Law. 1986;14:105-22.

Yung AR, McGorry PD, McFarlane CA, Jackson HJ, Patton GC, Rakkar A.
Monitoring and care of young people at incipient risk of psychosis. Schizophrenia
Bulletin. 1996;22:283-303.
Yung AR, Organ BA, Harris MG. Management of early psychosis in a generic adult
mental health service. The Australian and New Zealand Journal of Psychiatry.
2003;37:429-36.
Yung AR, Phillips LJ, Nelson B, Francey SM, PanYuen H, Simmons MB, et al.
Randomized controlled trial of interventions for young people at ultra high risk for
psychosis: 6-month analysis. Journal of Clinical Psychiatry. 2011;72:430-40.
Yung AR, Yuen HP, Berger G, Francey S, Hung TC, Nelson B, et al. Declining
transition rate in ultra high risk (prodromal) services: dilution or reduction of risk?
Zhu B, Ascher-Svanum H, Shi L, Faries D, Montgomery W, Marder SR. Time to

discontinuation of depot and oral first-generation antipsychotics in the usual care of
Ziegenbein M, Wittmann G, Kropp S. Aripiprazole augmentation of clozapine in

treatment-resistant schizophrenia: a clinical observation. Clinical drug investigation.
2006;26:117-24.
Zubin J, Spring B. Vulnerability--a new view of schizophrenia. Journal of Abnormal

Psychology. 1977;86:103-26.
Zullino DF, Delessert D, Eap CB, Preisig M, Baumann P. Tobacco and cannabis
smoking cessation can lead to intoxication with clozapine or olanzapine.
International Clinical Psychopharmacology. 2002;17:141-3.

Full Guideline 490503565

Uploaded by

Copyright:

Available Formats

Full Guideline 490503565

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Full Guideline 490503565

Uploaded by

Copyright:

Available Formats

“The original NICE schizophrenia guideline was of remarkable

superiority in its methodological quality compared with other

This guideline on Schizophrenia, commissioned by NICE and developed by the

An accompanying CD contains further information about the evidence, including:

UPDATED EDITION 2014 14

Treatment and management

National Clinical Guideline Number 178

National Collaborating Centre for Mental Health

Psychosis and schizophrenia in adults 1

Tim Kendall (Facilitator, Guideline Development Group)

Amina Yesufu Udechuku

Lucy Rebecca Burt

Psychosis and schizophrenia in adults 2

Psychosis and schizophrenia in adults 3

2 Psychosis and schizophrenia in adults ................................................................................ 14

3 Methods used to develop this guideline ............................................................................. 43

4 Carers’ experience .................................................................................................................... 65

Psychosis and schizophrenia in adults 4

6 Access and engagement......................................................................................................... 138

7 Interventions to promote physical health in adults......................................................... 157

8 Peer-provided and self-management interventions ........................................................ 184

9 Psychological therapy and psychosocial interventions .................................................. 208

10 Pharmacological interventions in the treatment and management of schizophrenia.....

Psychosis and schizophrenia in adults 5

11 Economic model - cost effectiveness of pharmacological interventions for people

12 Teams and service-level interventions ............................................................................... 457

13 Vocational rehabilitation ...................................................................................................... 531

14 Summary of recommendations ............................................................................................ 581

15 References ................................................................................................................................ 601

Psychosis and schizophrenia in adults 6

Victoria Bird, King’s College London

Brynmor Lloyd-Evans, Mental Health Sciences Unit, University College London

Alyssa Milton, Mental Health Sciences Unit, University College London

Daniel Tsoi, University of Sheffield

Sophia Winterbourne, London School of Economics

Those who conducted a review on behalf of the GDG:

Brynmor Lloyd-Evans, Mental Health Sciences Unit, University College London

Alyssa Milton, Mental Health Sciences Unit, University College London

Psychosis and schizophrenia in adults 7

Psychosis and schizophrenia in adults 8

• provide up-to-date evidence-based recommendations for the management of

1.1.2 Uses and limitation of clinical guidelines

Psychosis and schizophrenia in adults 9

In using guidelines, it is important to remember that the absence of empirical

1.1.3 Why develop national guidelines?

1.1.4 From national clinical guidelines to local protocols

Psychosis and schizophrenia in adults 10

1.1.5 Auditing the implementation of clinical guidelines

1.2 THE NATIONAL PSYCHOSIS AND SCHIZOPHRENIA

Psychosis and schizophrenia in adults 11

1.2.2 For whom is this guideline intended?

1.2.3 Specific aims of this guideline

1.2.4 The structure of this guideline

Psychosis and schizophrenia in adults 12

Text Box 1: Appendices in a separate file

2014 Search strategies for the identification of clinical studies Appendix 13

Psychosis and schizophrenia in adults 13

2.1 THE DISORDER

Typically, there will be a ‘prodromal’ period often characterised by some

Psychosis and schizophrenia in adults 14