• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Collaborating Centre for Mental Health (UK). Schizophrenia: Core Interventions in the Treatment and Management of Schizophrenia in Primary and Secondary Care (Update) [Internet]. Leicester (UK): British Psychological Society; 2009 Mar. (NICE Clinical Guidelines, No. 82.)

Cover of Schizophrenia

Schizophrenia: Core Interventions in the Treatment and Management of Schizophrenia in Primary and Secondary Care (Update) [Internet].

Show details

8Psychological therapy and psychosocial interventions in the treatment and management of schizophrenia

For the guideline update, all sections of the previous psychology chapter were updated, including the following psychological therapies and psychosocial intervention evidence reviews:

In addition, new reviews were conducted for the following interventions:

8.1. Introduction

Psychological therapies and psychosocial interventions have gained momentum in the treatment of schizophrenia over the past three 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. Secondly, 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 multi-disciplinary formulation and reflective practice, particularly where psychologists and allied mental health professionals operate within multidisciplinary teams (BPS, 2007). The ‘New Ways of Working’ report (BPS, 2007) also 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 in order to increase the workforce competent in the provision of psychological therapy. This chapter addresses the evidence base for the application of psychological and psychosocial treatments, generally in combination with antipsychotic medication for individuals, groups and families.

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 (Zubin & Spring, 1977; Nuechterlein, 1987) 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 treatments 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.

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.

Aims of psychological therapy and psychosocial interventions

The goals 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 the quality of life; reduce risk; improve communication and coping skills; and/or enhance treatment adherence. Research into psychological interventions needs to address a wide range of outcomes, as far as possible.

Therapeutic approaches identified

The following psychological therapies and psychosocial interventions were reviewed:

  • adherence therapy
  • arts therapies
  • cognitive remediation
  • cognitive–behavioural therapy (CBT)
  • counselling and supportive psychotherapy
  • family intervention
  • psychoanalysis and psychoanalytic/psychodynamic psychotherapy
  • psychoeducation
  • social skills training.

Multimodal interventions

Some researchers have combined two of the psychological and/or psychosocial interventions in order 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 which can be analysed together. Therefore multi-modal interventions (which 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, testing the effect, if any, of removing these multi-modal interventions. Where papers reported more than two treatment arms (for example, family intervention only vs. social skills training only vs. family intervention + social skills training), only data from the single intervention arms was entered into the appropriate analysis (for example, family intervention only vs. social skills training only). Papers assessing the efficacy of psychological treatments as adjuncts to discrete treatments outside the scope of the present update (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 previous guideline can be classed as multi-modal treatments as they systematically combine elements such as family intervention, social skills training and CBT for example, 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 multimodal, 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 achieved by a single intervention alone.

In addition, multimodal treatments could provide specific interventions, either concurrently or consecutively.

Competence to deliver psychological therapies

For the purpose of implementing the current guidelines, 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 intervention13.

8.2. Adherence therapy

8.2.1. Introduction

Pharmacological interventions have been the mainstay of treatment since their introduction in the 1950s, however, about 50% of people with schizophrenia and schizophrenic-form disorder are believed to be non-adherent (or non-compliant) to their medication (Nosěet al., 2003). It is estimated that non-adherence to medication leads to a higher relapse rate, repeated hospital admissions and thus increased economic and social burden for the service users themselves as well as for the mental health service (Gray et al., 2006; Robinson et al., 1999)14.

Against this background, ’compliance therapy’ was first developed by Kemp and colleagues targeting 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 & 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 roughly from thirty minutes to an hour (Kemp et al., 1996; Gray et al., 2006). 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
  • 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, in order to improve their adherence to medication and/or with the specific aim of improving symptoms, QoL and preventing relapse.

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

8.2.2. Clinical review protocol (adherence therapy)

The review protocol, including information about the databases searched and the eligibility criteria can be found in Table 52. The primary clinical questions can be found in Text box 2. A new systematic search for relevant studies was conducted for the guideline update. 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 8).

Table 52. Clinical review protocol for the review of adherence therapy.

Table 52

Clinical review protocol for the review of adherence therapy.

Box Icon

Text Box 2

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 to alternative management strategies (more...)

8.2.3. Studies considered for review15

Five RCTs (N = 649) met the inclusion criteria for the update. Although broadly based on a cognitive behavioural approach, KEMP1996 was reclassified as an adherence therapy paper as 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 15).

8.2.4. Adherence therapy versus control

For the update, five RCTs of adherence therapy versus any type of control were included in the meta-analysis (see Table 53 for a summary of the study characteristics). An evidence summary table for the comparison can be found in Chapter 10.

Table 53. Summary of study characteristics for adherence therapy.

Table 53

Summary of study characteristics for adherence therapy.

8.2.5. Clinical evidence summary

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.

8.2.6. 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, hospitalised for psychosis. The time horizon of the economic analysis was 18 months (RCT period plus naturalistic follow-up). Costs consisted of costs to the NHS (inpatient, outpatient, day-hospital care, A & E services, primary and community care) and criminal justice system costs incurred by arrests, court appearances, probation etc. Outcomes included relapse rates, BPRS and GAF scores, Drug Attitudes 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 time points of follow-up. 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 counselling in 1995/96 prices (p=0.92). Due to 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, as it was more effective than supportive counselling at a similar cost.

Details on the methods used for the systematic search of the economic literature are described in chapter 3. 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 14.

8.2.7. From evidence to recommendations

The current review found no consistent evidence to suggest that adherence therapy is effective in improving the critical outcomes of schizophrenia when compared to any other control. Although one UK-based study (KEMP1996) reported posi tive 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 in acute psychosis in the UK, as it was more effective than its comparator at a similar total cost. In addition to the aforementioned limitations of the KEMP1996 study, due to high attrition rates, the sample was very small to establish such a hypothesis. Based on the limited HE 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.

8.2.8. Recommendations

8.2.8.1.

Do not offer adherence therapy (as a specific intervention) to people with schizophrenia.

8.3. Arts therapies

8.3.1. Introduction

The arts therapy professions in the US and Europe have their roots in late nineteenth and early twentieth 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 forties which led to the emergence of formal trainings and professional bodies for art therapy, music therapy, dramatherapy and dance movement therapy. These treatments were further developed in psychiatric settings in the latter half of the 20th century (Bunt, 1994; Wood, 1997).

Whilst 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 below Psychodynamic and Psychoanalytic Therapies) tend to predominate in the UK (Crawford & Patterson, 2007).

More recently, arts therapy approaches to working with people with psychosis have begun to be more clearly defined, taking into consideration the phase and symptomatology of the illness (Gilroy& McNeilly, 2000; Jones, 1996). The arts therapy approaches 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, 1987; Rohricht & Priebe, 2006; Talwar et al., 2006; Ulrich,, 2007, Yang et al., 1998).

Art, music drama therapists and dance movement therapists16 practising in the UK are state registered professions, regulated by the Health Professions Council which requires specialist training at Masters level.

Definition

Arts therapies are complex interventions which 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 patient’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, Dramatherapy, and Music Therapy.

8.3.2. Clinical review protocol (Arts therapies)

The review protocol, including information about the databases searched and the eligibility criteria can be found in Table 54. The primary clinical questions can be found in Text box 2. A new systematic search for relevant RCTs was conducted for the guideline update (further information about the search strategy can be found in Appendix 8).

Table 54. Clinical review protocol for the review of arts therapies.

Table 54

Clinical review protocol for the review of arts therapies.

8.3.3. Studies considered for review

Seven RCTs (N = 406) met the inclusion criteria for the update. 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 15).

8.3.4. Arts therapies versus control

For the update, of the seven RCTs, six were included in the meta-analysis of arts therapy versus any type of control (see Table 55 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. An evidence summary table for each comparison can be found in Chapter 10.

Table 55. Summary of study characteristics for arts therapies.

Table 55

Summary of study characteristics for arts therapies.

8.3.5. Clinical evidence summary

The review found consistent evidence that arts therapies are effective in reducing negative symptoms when compared to 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 six 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, movement or art), and that arts therapies were equally as effective in reducing negative symptoms in both inpatient and outpatient populations.

8.3.6. 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 methods used for the systematic search of the economic literature are described in Chapter 3.

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 6–8 people in the UK, and have an average duration of one hour.

Arts therapies are provided by therapists with a specialist training at Masters level. The unit cost of a therapist providing arts therapies is not available. The salary scale of an arts therapist is lying across Bands 7 and 8a. This 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 as 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 £100–135 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 Health-Related Quality of Life (HRQoL) of people with schizophrenia by 0.005–0.007 annually, on a scale 0 (death) – 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 at 0.003–0.004 annually.

8.3.7. From evidence to recommendations

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-£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 0–1), then they would be cost-effective according to 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, as arts therapies is the only intervention that has been 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 across all trials. Consequently the GDG recommend 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 at present, arts therapies are the only interventions, both psychological and pharmacological, to demonstrate consistent efficacy in the reduction of negative symptoms. This, taken with the economic analysis, has led to the following recommendations.

8.3.8. Recommendations

Treatment of acute episode

8.3.8.1.

Consider offering arts therapies to all people with schizophrenia, particularly for the alleviation of negative symptoms. This can be started either during the acute phase or later, including in inpatient settings.

8.3.8.2.

Arts therapies should be provided by a Health Professions Council (HPC) registered arts therapist, with previous experience of working with people with 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 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.

Promoting recovery

8.3.8.3.

Consider offering arts therapies to assist in promoting recovery, particularly in people with negative symptoms.

8.3.9. Research recommendations

8.3.9.1.

An adequately powered randomised controlled trial should be conducted to investigate the clinical and cost-effectiveness of arts therapies compared to an active control (for example, sham music therapy) in people with schizophrenia.

8.3.9.2.

An adequately powered randomised controlled trial should be conducted to investigate the most appropriate duration and number of sessions for arts therapies in people with schizophrenia.

8.4. Cognitive–behavioural therapy

8.4.1. Introduction

Cognitive Behavioural Therapy is based on the premise that there is a relationship between thoughts, feelings and behaviour. Although Albert Ellis developed the first Cognitive Behavioural Therapy in the 1960s which he called Rational Emotive Behaviour Therapy, most Cognitive Therapy 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 post traumatic stress disorder. In the early 1990s, on the back of an increased understanding of the cognitive psychology of psychotic symptoms (Slade & Bentall, 1988; Frith, 1992; Garety & Hemsley, 1994), 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 focussing on the treatment of first episode psychosis (Jackson et al. 2005Jackson et al. 2007)) and people with schizophrenia and co-morbid substance use disorders (Barrowclough et al. 2001).

Definition

Cognitive-behavioural therapy was defined as a discrete psychological intervention where:

  • recipients establish links between their thoughts, feelings or actions with respect to the current or past symptoms, and/or functioning; and
  • the re-evaluation of their perceptions, beliefs or reasoning relate to the target symptoms.

In addition, a further component of the intervention should involve the following:

  • recipients monitor their own thoughts, feelings or behaviours with respect to the symptom or recurrence of symptoms; and/or
  • the promotion of alternative ways of coping with the target symptom; and/or
  • the reduction of distress; and/or
  • the improvement of functioning.

8.4.2. Clinical review protocol (cognitive-behavioural therapy)

The review protocol, including information about the databases searched and the eligibility criteria can be found in Table 56. The primary clinical questions can be found in Text box 2. For the guideline update, a new systematic search was conducted for relevant RCTs published since the previous guideline (further information about the search strategy can be found in Appendix 8 and information about the search for health economic evidence can be found in section 8.4.8).

Table 56. Clinical review protocol for the review of cognitive-behavioural therapy.

Table 56

Clinical review protocol for the review of cognitive-behavioural therapy.

8.4.3. Studies considered for review

In the previous guideline, 13 RCTs (N = 1,297) of CBT were included. One RCT from the previous guideline (KEMP1996) was removed from the update analysis and re-classified by the GDG as adherence therapy and a further three studies were removed due to inadequate numbers of participants (Garety1994; Levine1996; Turkington2000). The update search identified six papers providing follow-up data to existing RCTs, and 22 new RCTs, including those with CBT as part of a multimodal 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 15).

8.4.4. Cognitive-behavioural therapy versus control

For the update, 31 RCTs of CBT versus any type of control were included in the meta-analysis (see Table 57 for a summary of the study characteristics). However, this comparison was only used for outcomes in which there were insufficient studies to allow for 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. An evidence summary table for each comparison can be found in Chapter 10.

Table 57. Summary of study characteristics for CBT.

Table 57

Summary of study characteristics for CBT.

In addition to the primary analyses, sub-group analyses were used to explore certain characteristics of the trials17 (see Table 58 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, seven in the analysis comparing CBT with any control in participants experiencing an acute-episode, and 11 in the analysis comparing CBT with any control in participants during the promoting recovery phase. Finally, six RCTs were included in the analysis comparing group CBT with any control, and 18 were included in the analysis comparing individual CBT with any control. Multimodal trials were not included in the sub-group analyses. An evidence summary table for each comparison can be found in Chapter 10.

Table 58. Summary of study characteristics for CBT subgroup analyses.

Table 58

Summary of study characteristics for CBT subgroup analyses.

8.4.5. 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 CBT for psychosis (CBTp) trials this is unsurprising since 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), masters 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.

Competence does not appear to be directly correlated with training and a number of additional variables play a part. The Durham et al. (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 to demonstrate competence should be. Moreover, Durham et al. (2003) used two of these scales in their trial and found that therapy ratings did not correlate.

With regards 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.

8.4.6. 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 Turkington2002 study. The sub-group analysis indicated a higher drop-out rate amongst both black and ethnic minority groups. Additionally, compared to their white counterparts, the black and minority ethnic participants demonstrated significantly smaller changes in insight. Although 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 three and five percent of the sample respectively. With regards to the other studies included in the review, there was a paucity of information on the ethnicity of participants. Due to 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.

8.4.7. Clinical evidence summary

The review found consistent evidence that when compared to 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 previous 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 to both standard care and other active treatments. Furthermore, when compared to 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 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. Sub-group analyses also demonstrated additional effects for people with schizophrenia in the promoting recovery phase both with and without persistent symptoms. In particular, when compared to 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.

8.4.8. Health economic evidence

Systematic literature review

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 Chapter 3. 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 14.

Kuipers and colleagues (1998) evaluated the cost-effectiveness of CBT added to standard care compared to standard care alone in 60 people with medication-resistant psychosis participating in a 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 and day-hospital care, 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 aspect, 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 and £1,403 for CBT added to standard care and standard care alone, respectively (p=0.416, 1996 prices). The study had insufficient 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 nine 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 (TAU) versus TAU alone in 90 people hospitalised for an acute psychotic episode participating in a 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 care, primary and community care, medication and residential care. 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 number 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 Band 6b and 8. This is comparable to the salary level of a clinical psychologist. Therefore, the unit cost of clinical psychologists was used in order to estimate an average intervention cost. The unit cost of a clinical psychologist has been estimated at £67 per hour of client contact in 2006/07 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 as 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, 2006/07 price). 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 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 2006/7 prices.

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 (Hospital Episode Statistics; NHS, The Information Centre, 2008A) 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; DH, 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, 2008A). 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 underestimate 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 (DH, 2008), 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 2006/07 prices.

Clinical data on hospitalisation rates following provision of CBT

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 to people receiving standard care alone. Table 59 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, as well as the results of the meta-analysis.

Table 59. Studies considered in the economic analysis of CBT in addition to standard care versus standard care alone and results of meta-analysis.

Table 59

Studies considered in the economic analysis of CBT in addition to standard care versus standard care alone and results of 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 (relative risk 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% confidence intervals of relative risk: 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 15. The forest plots of the respective meta-analysis are provided in Appendix 16.

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% confidence intervals (CIs) of the relative risk (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 conduct of the study 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, 2008A) 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 60.

Table 60. Results of cost analysis comparing CBT in addition to standard care versus standard care alone per person with schizophrenia.

Table 60

Results of cost analysis comparing CBT in addition to standard care versus standard care alone per person with schizophrenia.

Sensitivity analysis

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 due to 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 61.

Table 61. Results of sensitivity analysis of offering CBT in addition to standard care to people with schizophrenia.

Table 61

Results of sensitivity analysis of offering CBT in addition to standard care to people with schizophrenia.

Discussion

The economic analysis showed that CBT is likely to be an overall cost-saving intervention for people with schizophrenia as 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, respectively, as well as broader financial implications to the society (for example, potential increased productivity) associated with 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.

8.4.9. From evidence to recommendations

The conclusions drawn in the previous guideline regarding the efficacy of CBT have been supported by the updated 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 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 previous 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 previous 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 sub-group analyses assessing the impact of treatment duration and number of sessions, they did note that the evidence for CBT is primarily driven by studies which included at least 16 planned sessions. Thus to incorporate the current state of evidence and based upon expert consensus, the GDG modified the previous 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 feel that this is an important area for future development and have subsequently included 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 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 previous recommendations regarding the provision of CBT to people with schizophrenia.

8.4.10. Recommendations

8.4.10.1.

Offer cognitive behavioural therapy (CBT) to all people with schizophrenia. This can be started either during the acute phase18 or later, including in inpatient settings.

How to deliver psychological interventions

8.4.10.2.

CBT should be delivered on a one-to-one basis over at least 16 planned sessions and:

  • follow a treatment manual19 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.

Promoting recovery

8.4.10.3.

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 8.4.10.2.

8.4.11. Research recommendations

8.4.11.1.

An adequately powered randomised controlled trial should be conducted to investigate the most appropriate duration and number of sessions for CBT in people with schizophrenia.

8.4.11.2.

An adequately powered randomised controlled trial should be conducted to investigate CBT delivered by highly trained therapists and mental health professionals compared to brief training of therapists in people with schizophrenia.

8.4.11.3.

Research is needed to identify the competencies required to deliver effective CBT to people with schizophrenia.

8.5. Cognitive remediation

8.5.1. Introduction

The presence of cognitive impairment in a proportion of people with schizophrenia has been recognised since the term 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; Neuchterlein 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 6). 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 stated improved functioning as an additional aim (Wykes1999 & 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).

Since the use of these methods in the treatment of schizophrenia is still developing, and early studies had mixed results (Pilling et al., 2002b) there remains uncertainty as to which techniques should be used (Wykes1999 & 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 intervention updated within the current review. In these cases the intervention has been classified as multi-modal intervention and subjected to sensitivity analyses (see introduction).

Definition

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
  • the procedure is implemented with 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.

8.5.2. Clinical review protocol (cognitive remediation)

The review protocol, including information about the databases searched and the eligibility criteria can be found in Table 62. The primary clinical questions can be found in Text box 2. For the guideline update, a new systematic search was conducted for relevant RCTs published since the previous guideline (further information about the search strategy can be found in Appendix 8) 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 being that only sustained improvement would be likely to have an impact on other critical outcomes, such as mental state, psychosocial functioning, hospitalisation and relapse

Table 62. Clinical review protocol for the review of cognitive remediation.

Table 62

Clinical review protocol for the review of cognitive remediation.

8.5.3. Studies considered for review

In the previous guideline, seven RCTs of cognitive remediation were included. Two trials (Bellack2001 and Tompkins1995) were removed from the update analysis as the GDG felt that they did not meet the definition of CRT. The update search identified three 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 trial employed a variety of different methods and in some cases applied cognitive remediation in combination with a variety of other psychological or psychosocial interventions20. In total, 25 trials (N = 1,390) met inclusion criteria. All of the trials were published in peer-reviewed journals between 1994 and 2008 (further information about both included and excluded studies can be found in Appendix 15).

8.5.4. Cognitive remediation versus control

For the update, 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 62. Consequently, 20 RCTs of cognitive remediation versus any type of control were included in the meta-analysis (see Table 63 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. An evidence summary table for each comparison can be found in Chapter 10.

Table 63. Summary of study characteristics for cognitive remediation.

Table 63

Summary of study characteristics for cognitive remediation.

8.5.5. Clinical evidence summary

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 to standard care may improve social functioning. However, this effect was driven by a range of studies conducted by Velligan and colleagues (VELLIGAN2000VELLIGAN2002VELLIGAN2008A, 2008B), in which the intervention was more 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, though well-conducted, did not report evidence of improvement in social or vocational functioning or symptoms at either end of treatment or follow-up.

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 over estimated for all outcomes.

8.5.6. From evidence to recommendations

The previous 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 et al. (2007) review suggested positive effects for symptoms and functioning, this may in part, be attributed to the fact that their review included a number of studies which failed to meet the inclusion criteria set out by the GDG (for example, minimum number of participants, CRT 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 did comprehensively report outcomes at both end 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 previous recommendation.

The GDG did however note 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 guideline update 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.

8.5.7. Research recommendations

8.5.7.1.

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

8.5.7.2.

An adequately powered randomised controlled trial with longer-term follow-up should be conducted to investigate the clinical and cost-effectiveness of vocational rehabilitation plus cognitive remediation compared to vocational rehabilitation alone in people with schizophrenia.

8.6. Counselling and supportive therapy

8.6.1. Introduction

In the 1950s, Carl Rogers under the influence of Adler and Rank, a pioneering US psychologist influenced by Adler and Rank Carl Rogers devised ‘client-centred’ and later ‘person-centred’ counselling. This was a reaction against behaviourist and psychodynamic schools which 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). Since then, 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 & Hertz 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 kind 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) report that a patient’s feeling of being listened to and understood is a strong predictor of medication compliance for example. Also, according to McCabe and Priebe (2004) the therapeutic relationship is a reliable predictor of patient outcome in mainstream psychiatric care.

Definition

Counselling or supportive therapy was defined as a discrete psychological intervention where:

  • the intervention is facilitative, non-directive and/or relationship focused, with the content largely determined by the service user; and
  • the intervention does not fulfil the criteria for any other psychological intervention.

8.6.2. Clinical review protocol (counselling and supportive therapy)

The review protocol, including information about the databases searched and the eligibility criteria used for this section of the guideline can be found in Table 64. The primary clinical questions can be found in Text box 2. A new systematic search for relevant RCTs, published since the previous guideline, was conducted for the guideline update (further information about the search strategy can be found in Appendix 8).

Table 64. Clinical review protocol for the review of counselling and supportive therapy.

Table 64

Clinical review protocol for the review of counselling and supportive therapy.

8.6.3. Studies considered for review

In the previous guideline, 14 RCTs (N = 1,143) of counselling and supportive therapy were included. Two studies included in the previous guideline (Levine1998; Turkington2000) were excluded from the update due to inadequate numbers of participants. The update search identified four papers providing follow-up data to existing trials, and six new trials. In total, 18 RCTs (N = 1,610) met inclusion criteria for the update. All were published in peer-reviewed journals between 1973 and 2007 (further information about both included and excluded studies can be found in Appendix 15).

8.6.4. Counselling and supportive therapy versus control

For the update, 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 CBT21 (see Table 65 for a summary of the study characteristics). An evidence summary table for each comparison can be found in Chapter 10.

Table 65. Summary of study characteristics for counselling and supportive therapy.

Table 65

Summary of study characteristics for counselling and supportive therapy.

8.6.5. Clinical evidence summary

In 17 RCTs including 1,586 participants, there was evidence to suggest that counselling and supportive psychotherapy does not improve outcomes in schizophrenia, when compared with standard care and other active treatments, most notably CBT. A sub-group analysis of counselling and supportive therapy versus CBT favoured the CBT intervention for a number of outcomes including relapse. However, it must be noted that in these studies, counselling and supportive therapy was used as a comparator to control primarily for therapist time and attention and thus was not the focus of the research.

8.6.6. From evidence to recommendations

In the previous 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 this update does not justify changing this recommendation. The GDG do 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.

8.6.7. Recommendations

8.6.7.1.

Do not routinely offer counselling and supportive psychotherapy (as specific interventions) to people with 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.

8.7. Family intervention

8.7.1. 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 others (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 and 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 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 other of the following components:
  • problem solving/crisis management work, or
  • intervention with the identified service user.

8.7.2. Clinical review protocol (family intervention)

The review protocol, including information about the databases searched and the eligibility criteria used for this section of the guideline can be found in Table 66. The primary clinical questions can be found in Text box 2. A new systematic search for relevant RCTs, published since the previous guideline, was conducted for the guideline update (further information about the search strategy can be found in Appendix 8 and information about the search for health economic evidence can be found in section 8.7.8).

Table 66. Clinical review protocol for the review of family intervention.

Table 66

Clinical review protocol for the review of family intervention.

8.7.3. Studies considered for review

In the previous guideline, 18 RCTs (N = 1,458) of family intervention were included. One study (Posner1992) included in the previous guideline was re-classified as Psychoeducation for the update and two previous trials were classified as having family intervention as part of a multimodal treatment (Herz2000 and Lukoff1986). The update search 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 update. All were published in peer-reviewed journals between 1978 and 2008 (further information about both included and excluded studies can be found in Appendix 15).

8.7.4. Family intervention versus control

For the update, one of the included studies (CHENG2005) did not provide useable data for any of the critical outcomes listed in Table 66, 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, 8 compared family intervention with other active treatments. Additionally, 5 trials directly compared a multiple family intervention with a single family intervention (see Table 67 for a summary of the study characteristics). An evidence summary table for each comparison can be found in Chapter 10.

Table 67. Summary of study characteristics for family intervention.

Table 67

Summary of study characteristics for family intervention.

Sub-group analyses were also used to examine whether the format of the family intervention had an impact on outcome (10 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 sub-group 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 intervention22 (see Table 68 for a summary of the studies included in each sub-group comparison).

Table 68. Summary of study characteristics for family intervention subgroup comparisons.

Table 68

Summary of study characteristics for family intervention subgroup comparisons.

8.7.5. 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.

8.7.6. Ethnicity

Although the data on ethnicity was limited, a sub-group analysis looking at the efficacy of family intervention in an ethnically diverse population was conducted (See Chapter 5 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, which included the use of language matching and ethnospecific 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 were conducted with BME participants from the UK; therefore the generalisability of such findings is limited. Furthermore, at present little research exists which directly compares the efficacy and acceptability of culturally and non-culturally modified approaches.

8.7.7. Clinical evidence summary

In 32 RCTs including 2,429 participants, there was robust and consistent evidence for the efficacy of family intervention. When compared to standard care or any other control, there was a reduction in the risk of relapse with NNTs of 4 (95% CIs 3.23 to 5.88) and 6 (95% CIs 3.85 to 9.09) at the end of treatment and up to 12 months following treatment, respectively. 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 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 sub-group analyses conducted for the update 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, sub-group comparisons which indirectly compared single to multiple family intervention, demonstrated some limited evidence to suggest that only the former may be efficacious in reducing hospital admission.

8.7.8. Health economic evidence

Systematic literature review

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 previous NICE schizophrenia 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 guideline update are described in Chapter 3; details on the respective methods in the previous NICE schizophrenia guideline are provided in Appendix 17. The following text is derived from the previous schizophrenia guideline:

“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; Tarrier et al., 1991; Leff et al., 2001), while two studies were economic evaluations (Liberman et al., 1987; McFarlane et al., 1995). Of these, two economic analyses were conducted in the UK (Tarrier et al., 1991; Leff et al., 2001), 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 to 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 health care costs offset the extra treatment costs. One study (Leff et al., 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.”

The evidence table for the above studies as appeared in the previous schizophrenia guideline are included in Appendix 14.

Economic modelling

Objective

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 a main analysis, irrespective of the mode of delivery, as it was difficult to distinguish between single and multi programmes: the majority of studies described family intervention programmes that were predominantly single or multi, but might have some multi or single component, respectively; some of the interventions combined equally single and multi sessions.

Apart from the main meta-analysis, studies of family interventions 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) multi-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 multi-family intervention was not associated with a statistically significant respective effect. On the other hand, single- and multi-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) multi-family intervention. Meta-analysis of these studies showed that single and multi family intervention had no significant difference in clinical outcomes. However, participants showed a clear preference for single interventions, as expressed in drop-out 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 multi-family intervention the intervention cost is spread in more than one families.

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 GDG expert opinion on 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 2 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 to the salary level of a clinical psychologist. Therefore, the unit cost of clinical psychologist was used in order to estimate an average intervention cost. The unit cost of a clinical psychologist has been estimated at £67 per hour of client contact in 2006/07 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 as the latter are not available for clinical psychologists.

Based on the above resource use estimates and the unit cost of clinical psychologist, the cost of providing a full course of family intervention was estimated at £2,680 per person with schizophrenia in 2006/7 prices.

Costs of hospitalisation/cost-savings from reduction in hospitalisation rates As described in section 8.4.8, 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 (DH, 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, definition of relapse was more consistent across studies (and countries). For this reason, for the economic analysis, it was decided to use pooled data on relapse rather hospitalisation rates; these data would be used, subsequently, to estimate hospitalisation rates relevant to people with schizophrenia in the UK, in order to calculate cost-savings from reduction in 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 4–24 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 separately relapse data during treatment from respective data after treatment, but other studies included events that occurred during treatment in the reported follow-up data. Therefore, taking into account both sets of data might 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, as these data were homogeneous and referred to events that occurred within the same study phase. It is acknowledged, though, that the cost-savings estimated using exclusively data reported during treatment are probably underestimates of the true cost-savings, since the beneficial effect of family intervention on relapse remains for a substantial period after completion of treatment.

Table 69 shows the family intervention studies included in the meta-analysis of relapse rate data for 1–12 months into treatment, the relapse rates for each treatment arm reported in the individual studies, as well as the results of the meta-analysis.

Table 69. Studies considered in the economic analysis of family intervention added to standard care versus standard care alone and results of meta-analysis (1–12 months into treatment).

Table 69

Studies considered in the economic analysis of family intervention added to standard care versus standard care alone and results of meta-analysis (1–12 months into treatment).

The results of meta-analysis show that family intervention, when added to standard care, reduces the rate of relapse in people with schizophrenia during the intervention period (relative risk of relapse of family intervention added to standard care versus standard care alone: 0.52). This result was significant at the 0.05 level (95% confidence intervals of relative risk: 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 relative risk up to 24 months following provision of family intervention 0.63 with 95% confidence intervals 0.52 to 0.78).

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 relative risk 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 15. The forest plots of the respective meta-analysis are provided in Appendix 16.

Association between relapse and hospitalisation rates

In the UK, people with schizophrenia experiencing a relapse are mainly treated either as inpatients, or by Crisis Resolution and Home Treatment Teams (CRHTT). Glover and colleagues (2006) examined the reduction in hospital admission rates in England, following implementation of CRHTT. They reported that 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 in hospital, and the rest 22.7% would be seen by CRHTT.

Sensitivity analysis

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% confidence intervals (CIs) of the relative risk (RR) of relapse of family intervention added to standard care versus standard care alone.
  • Change in the total number of hours of a family intervention course (20 in the base-case analysis) to a range between 15 and 25.
  • 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 introduction of CRHTT, which was 38.4% according to Glover and colleagues, 2006).
  • 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

Provision of 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 equalling £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 70.

Table 70. Results of cost analysis comparing family intervention in addition to standard care versus standard care alone per person with schizophrenia.

Table 70

Results of cost analysis comparing family intervention in addition to standard care versus standard care alone per person with schizophrenia.

Sensitivity analysis

The results of the base-case analysis were overall robust to 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, addition of family intervention to standard care resulted in overall cost-savings due to a substantial reduction in relapse and subsequent hospitalisation costs. The only scenario in which family intervention was not cost-saving but instead incurred 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 71.

Table 71. Results of sensitivity analysis of providing family intervention in addition to standard care for people with schizophrenia.

Table 71

Results of sensitivity analysis of providing family intervention in addition to standard care for people with schizophrenia.

Discussion

The economic analysis showed that family intervention for people with schizophrenia is likely to be an overall cost-saving intervention, as 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, then the net cost of providing family intervention was found to lie between −£1,326 (overall net saving) and £263 per person with schizophrenia.

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 CRHTT, 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 HRQoL of people with schizophrenia and their carers, following reduction in relapse rates, strengthens further the argument that family intervention is likely a cost-effective option for people with schizophrenia in the UK.

8.7.9. From evidence to recommendations

There was sufficient evidence in the previous 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 update continue to support the evidence demonstrated in the previous guideline with regards 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 to multiple family interventions, this must be considered in the context of current practice and 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 reduces relapse rates also 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 regards 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 from any conclusions or make any recommendations relating to practice. However, the GDG acknowledge 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 previous guideline. Although there was a lack of evidence for the use of culturally adapted family interventions within the UK, the GDG acknowledge that this is an important area warranting further investigation given the evidence previously discussed relating to inequality of access for people from BME groups (see Chapter 5).

8.7.10. Recommendations

Treatment of acute episode

8.7.10.1.

Offer family intervention to all families of people with schizophrenia who live with or are in close contact with the service user. This can be started either during the acute phase23 or later, including in inpatient settings.

8.7.10.2.

Family intervention should:

  • include the person with 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 schizophrenia
  • have a specific supportive, educational or treatment function and include negotiated problem solving or crisis management work.

Promoting recovery

8.7.10.3.

Offer family intervention to families of people with schizophrenia who live with or are in close contact with the service user. Deliver family intervention as described in recommendation 8.7.10.2.

8.7.10.4.

Family intervention may be particularly useful for families of people with schizophrenia who have:

  • recently relapsed or are at risk of relapse
  • persisting symptoms.

8.7.11. Research recommendations

8.7.11.1.

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

8.7.11.2.

Research is needed to identify the competencies required to deliver effective family intervention to people with schizophrenia and their carers.

8.8. Psychodynamic and Psychoanalytic therapies

8.8.1. Introduction

Psychoanalysis and its derivatives, often termed psychoanalytic and psychodynamic psychotherapies, originate from the work of Freud at 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, 1914, 1933). 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 the building of a relationship based on genuineness and warmth whilst maintaining optimal distance (Gabbard, 1994)

Randomised controlled trials were undertaken in the 1970s and 1980s to investigate the use of psychoanalytically oriented psychotherapy. Research into the effects of psychoanalytic approaches in the treatment of schizophrenia has been repeated more recently, with mixed results (Fenton & McGlashan, 1995; Jones et al., 1999; Mari & Streiner, 1999), leading to the publication of a Cochrane Review of the subject (Malmberg & Fenton, 2001).

Definition

Psychodynamic interventions were defined as:

  • regular therapy sessions based on a psychodynamic or psychoanalytic model, and
  • sessions could rely on a variety of strategies, including explorative insight oriented, supportive or directive activity, applied flexibly.

To be considered well-defined psychodynamic psychotherapy, the intervention needed to include working with transference and unconscious processes.

Psychoanalytic interventions were defined as:

  • regular, individual sessions planned to continue for at least one year; and
  • analysts were 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.

8.8.2. Clinical review protocol (psychodynamic and psychoanalytic therapies)

The review protocol, including information about the databases searched and the eligibility criteria used for this section of the guideline can be found in Table 72. The primary clinical questions can be found in Text box 2. A new systematic search for relevant RCTs, published since the previous guideline, was conducted for the guideline update (further information about the search strategy can be found in Appendix 8).

Table 72. Clinical review protocol for the review of psychodynamic and psychoanalytic therapies.

Table 72

Clinical review protocol for the review of psychodynamic and psychoanalytic therapies.

8.8.3. Studies considered for review

In the previous guideline, 3 RCTs (N = 492) of psychodynamic and psychoanalytic therapies were included. The update search identified one new trial. In total, four RCTs (N = 558) met the inclusion criteria for the update. All of the trials were published in peer-reviewed journals between 1972 and 2003. In addition, one study identified in the update search 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 8).

8.8.4. Psychodynamic and psychoanalytic therapies versus control

For the update, two RCTs of psychodynamic and psychoanalytic therapies versus any type of control were included in the meta-analysis. Additionally, two trials included in the previous guideline directly compared the format of the intervention, and one trial compared insight orientated versus reality adaptive therapy, and another trial compared individual versus group therapy24 (see Table 73 for a summary of the study characteristics). An evidence summary table for each comparison can be found in Chapter 10.

Table 73. Summary of study characteristics for psychodynamic and psychoanalytic therapies.

Table 73

Summary of study characteristics for psychodynamic and psychoanalytic therapies.

8.8.5. Clinical evidence summary

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

8.8.6. From evidence to recommendations

In the previous 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 update 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 did note, 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 to other interventions in research trials. At present, the GDG are not aware of any well conducted RCTs assessing the efficacy of newer contemporary forms of psychodynamic and psychoanalytic therapy. It is therefore view of the GDG that further well-conducted research is warranted.

8.8.7. Recommendations

8.8.7.1.

Healthcare professionals may consider using psychoanalytic and psychodynamic principles to help them understand the experiences of people with schizophrenia and their interpersonal relationships.

8.8.8. Research recommendations

8.8.8.1.

A pilot RCT should be conducted to assess the efficacy of contemporary forms of psychodynamic therapy when compared to standard care and other active psychological and psychosocial interventions.

8.9. Psychoeducation

8.9.1. 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 information provider and service user or family, which has the primary aim of offering information about the condition; and
  • service users or families may also be provided with support and management strategies.

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

8.9.2. Clinical review protocol (psychoeducation)

The review protocol, including information about the databases searched and the eligibility criteria used for this section of the guideline can be found in Table 74. The primary clinical questions can be found in Text box 2. A new systematic search for relevant RCTs, published since the previous guideline, was conducted for the guideline update (further information about the search strategy can be found in Appendix #).

Table 74. Clinical review protocol for the review of psychoeducation.

Table 74

Clinical review protocol for the review of psychoeducation.

8.9.3. Studies considered for review

In the previous guideline, 10 RCTs (N = 1,070) of psychoeducation were included. The update search identified three papers providing follow-up data to existing trials, and 10 new trials. In the previous guideline, one study (Posner1992) included in family intervention review was reclassified as psychoeducation for the update. In total, 21 trials (N = 2,016) met the inclusion criteria for the update. All were published in peer-reviewed journals between 1987 and 2008 (further information about both included and excluded studies can be found in Appendix 15).

8.9.4. Psychoeducation versus control

For the update, four of the included studies (Jones2001, SIBITZ2007, Smith1987, XIA>NG2007) only included a direct comparison of different types of Psychoeducation and one trial (AGARA2007) did not provide any useable data thus 16 trials of psychoeducation versus any type of control were included in the meta-analysis (see Table 75 for a summary of the study characteristics). Sub-group 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 treatment25). An evidence summary table for each comparison can be found in Chapter 10.

Table 75. Summary of study characteristics for psychoeducation.

Table 75

Summary of study characteristics for psychoeducation.

8.9.5. Clinical evidence summary

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 GDGs definition of psychoeducation.

8.9.6. From evidence to recommendations

In the previous 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 here 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 update does not justify making a recommendation. However, the GDG acknowledge the importance of the provision of good quality and accessible information to all people with schizophrenia and their carers, and have hence made a number of related recommendations (see 4.6.4.1, 4.6.5.2 and 5.3.10.1).

8.10. Social skills training

8.10.1. 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 et al., 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, sometimes also, 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 USA 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 to respond to and provide appropriate social reinforcement.

8.10.2. Clinical review protocol (social skills training)

A new systematic search for relevant RCTs, published since the previous guideline, was conducted for the guideline update. Information about the databases searched and the eligibility criteria used for this section of the guideline can be found in Table 76 (further information about the search strategy can be found in Appendix 8).

Table 76. Clinical review protocol for the review of social skills training.

Table 76

Clinical review protocol for the review of social skills training.

8.10.3. Studies considered for review

In the previous guideline, nine RCTs (N = 436) of social skills training were included. One RCT from the previous guideline (Finch1977) was removed from the update analysis due to inadequate numbers of participants and one RCT (Eckmann 1992) was reclassified as social skills training and included in the analysis. The update search identified 14 new trials. In total, 23 trials (N = 1,471) met the inclusion criteria for the update. All were published in peer-reviewed journals between 1983 and 2007 (further information about both included and excluded studies can be found in Appendix 15).

8.10.4. Social skills training versus control

For the update, 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 77 for a summary of the study characteristics). Sub-group analyses were used to examine the impact of the type of comparator26 (10 trials used standard care as the comparator, and 10 trials used another active treatment) An evidence summary table for each comparison can be found in Chapter 10.

Table 77. Summary of study characteristics for social skills training.

Table 77

Summary of study characteristics for social skills training.

8.10.5. Clinical evidence summary

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) with multiple methodological problems.

8.10.6. From evidence to recommendations

In the previous 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 recent review (Kurtz & Mueser. 2008) indicated effects for social functioning, symptom severity and relapse this may be attributed to the inclusion of a number of studies which 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 update does not justify changing the conclusions drawn in the previous guideline.

8.10.7. Recommendations

8.10.7.1.

Do not routinely offer social skills training (as a specific intervention) to people with schizophrenia.

8.11. Recommendations (across all treatments)27

8.11.1. Principles in the provision of psychological therapies

8.11.1.1.

When providing psychological interventions, routinely and systematically monitor a range of outcomes across relevant areas, including service user satisfaction and, if appropriate, carer satisfaction.

8.11.1.2.

Healthcare teams working with people with 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.
8.11.1.3.

Healthcare professionals providing psychological interventions should:

  • have an appropriate level of competence in delivering the intervention to people with schizophrenia
  • be regularly supervised during psychological therapy by a competent therapist and supervisor.
8.11.1.4.

Trusts should provide access to training that equips healthcare professionals with the competencies required to deliver the psychological therapy interventions recommended in this guideline.

8.11.1.5.

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.

Training and competency reviews are presented for only those recommended interventions.

Further information about medicines concordance and adherence to treatment can be found in the NICE guideline on this topic (see http://www​.nice.org.uk).

Here 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 15.

Registration pending

Existing sub-group 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 conlusions based on these sub-groups. Please refer to chapter 10 for the evidence tables for all sub-group comparisons conducted.

CBT should be delivered as described in recommendation 8.4.10.2.

Treatment manuals that have evidence for their efficacy from clinical trials are preferred.

Trials assessing the efficacy of cognitive remediation as an adjunct to non-psychological or psychosocial interventions, such as vocational rehabilitation programmes were outside the scope of the review (see introduction)

Existing sub-group comparisons exploring the format of the interention (group vs. individual sessions) was also updated. However, there was insufficient data to draw any conclusions based on this sub-group. Please refer to chapter 10 for the evidence tables for all sub-group comparisons conducted.

Existing sub-group comparisons exploring the country of the trial, the number of treatment sessions, and the family characteristics (High Emotional Expression vs. everything) were also updated. However, there was insufficient data to draw any conclusions based on these sub-groups. Please refer to chapter 10 for the evidence tables for all sub-group comparisons conducted.

Family intervention should be delivered as described in recommendation 8.7.10.2.

Existing sub-group comparing psychodynamic and psychoanalytic therapies to standard care and other active treatments and psychodynamic therapy to group psychodynamic therapy were also updated. However, there was insufficient data to draw any conclusions based on these sub-groups. Please refer to chapter 10 for the evidence tables for all sub-group comparisons conducted.

Existing sub-group 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 sub-groups. Please refer to chapter 10 for the evidence tables for all sub-group comparisons conducted.

Existing sub-group comparisons exploring the duration of treatment and treatment setting were also updated. However, there was insufficient data to draw any conclusions based on these sub-groups. Please refer to chapter 10 for the evidence tables for all sub-group comparisons conducted.

Recommendations for specific interventions can be found at the end of each review (see the beginning of this chapter for further information).

Footnotes

13

Training and competency reviews are presented for only those recommended interventions.

14

Further information about medicines concordance and adherence to treatment can be found in the NICE guideline on this topic (see http://www​.nice.org.uk).

15

Here 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 15.

16

Registration pending

17

Existing sub-group 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 conlusions based on these sub-groups. Please refer to chapter 10 for the evidence tables for all sub-group comparisons conducted.

18

CBT should be delivered as described in recommendation 8.4.10.2.

19

Treatment manuals that have evidence for their efficacy from clinical trials are preferred.

20

Trials assessing the efficacy of cognitive remediation as an adjunct to non-psychological or psychosocial interventions, such as vocational rehabilitation programmes were outside the scope of the review (see introduction)

21

Existing sub-group comparisons exploring the format of the interention (group vs. individual sessions) was also updated. However, there was insufficient data to draw any conclusions based on this sub-group. Please refer to chapter 10 for the evidence tables for all sub-group comparisons conducted.

22

Existing sub-group comparisons exploring the country of the trial, the number of treatment sessions, and the family characteristics (High Emotional Expression vs. everything) were also updated. However, there was insufficient data to draw any conclusions based on these sub-groups. Please refer to chapter 10 for the evidence tables for all sub-group comparisons conducted.

23

Family intervention should be delivered as described in recommendation 8.7.10.2.

24

Existing sub-group comparing psychodynamic and psychoanalytic therapies to standard care and other active treatments and psychodynamic therapy to group psychodynamic therapy were also updated. However, there was insufficient data to draw any conclusions based on these sub-groups. Please refer to chapter 10 for the evidence tables for all sub-group comparisons conducted.

25

Existing sub-group 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 sub-groups. Please refer to chapter 10 for the evidence tables for all sub-group comparisons conducted.

26

Existing sub-group comparisons exploring the duration of treatment and treatment setting were also updated. However, there was insufficient data to draw any conclusions based on these sub-groups. Please refer to chapter 10 for the evidence tables for all sub-group comparisons conducted.

27

Recommendations for specific interventions can be found at the end of each review (see the beginning of this chapter for further information).

Copyright © 2009, National Collaborating Centre for Mental Health.
Bookshelf ID: NBK11688
PubReader format: click here to try

Views

  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (3.1M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...