Early intervention for psychosis

Max Marshall; John Rathbone

doi:10.1002/14651858.CD004718.pub3

. Author manuscript; available in PMC: 2014 Sep 15.

Published in final edited form as: Cochrane Database Syst Rev. 2011 Jun 15;(6):CD004718. doi: 10.1002/14651858.CD004718.pub3

Early intervention for psychosis

Max Marshall ¹, John Rathbone ²

PMCID: PMC4163966 EMSID: EMS57372 PMID: 21678345

Abstract

Background

Proponents of early intervention have argued that outcomes might be improved if more therapeutic efforts were focused on the early stages of schizophrenia or on people with prodromal symptoms. Early intervention in schizophrenia has two elements that are distinct from standard care: early detection, and phase-specific treatment (phase-specific treatment is a psychological, social or physical treatment developed, or modified, specifically for use with people at an early stage of the illness).

Early detection and phase-specific treatment may both be offered as supplements to standard care, or may be provided through a specialised early intervention team. Early intervention is now well established as a therapeutic approach in America, Europe and Australasia.

Objectives

To evaluate the effects of: (a) early detection; (b) phase-specific treatments; and (c) specialised early intervention teams in the treatment of people with prodromal symptoms or first-episode psychosis.

Search methods

We searched the Cochrane Schizophrenia Group Trials Register (March 2009), inspected reference lists of all identified trials and reviews and contacted experts in the field.

Selection criteria

We included all randomised controlled trials (RCTs) designed to prevent progression to psychosis in people showing prodromal symptoms, or to improve outcome for people with first-episode psychosis. Eligible interventions, alone and in combination, included: early detection, phase-specific treatments, and care from specialised early intervention teams. We accepted cluster-randomised trials but excluded non-randomised trials.

Data collection and analysis

We reliably selected studies, quality rated them and extracted data. For dichotomous data, we estimated relative risks (RR), with the 95% confidence intervals (CI). Where possible, we calculated the number needed to treat/harm statistic (NNT/H) and used intention-to-treat analysis (ITT).

Main results

Studies were diverse, mostly small, undertaken by pioneering researchers and with many methodological limitations (18 RCTs, total n=1808). Mostly, meta-analyses were inappropriate. For the six studies addressing prevention of psychosis for people with prodromal symptoms, olanzapine seemed of little benefit (n=60, 1 RCT, RR conversion to psychosis 0.58 CI 0.3 to 1.2), and cognitive behavioural therapy (CBT) equally so (n=60, 1 RCT, RR conversion to psychosis 0.50 CI 0.2 to 1.7). A risperidone plus CBT plus specialised team did have benefit over specialist team alone at six months (n=59, 1 RCT, RR conversion to psychosis 0.27 CI 0.1 to 0.9, NNT 4 CI 2 to 20), but this was not seen by 12 months (n=59, 1 RCT, RR 0.54 CI 0.2 to 1.3). Omega 3 fatty acids (EPA) had advantage over placebo (n=76, 1 RCT, RR transition to psychosis 0.13 CI 0.02 to 1.0, NNT 6 CI 5 to 96). We know of no replications of this finding.

The remaining trials aimed to improve outcome in first-episode psychosis. Phase-specific CBT for suicidality seemed to have little effect, but the single study was small (n=56, 1 RCT, RR suicide 0.81 CI 0.05 to 12.26). Family therapy plus a specialised team in the Netherlands did not clearly affect relapse (n=76, RR 1.05 CI 0.4 to 3.0), but without the specialised team in China it may (n=83, 1 RCT, RR admitted to hospital 0.28 CI 0.1 to 0.6, NNT 3 CI 2 to 6). The largest and highest quality study compared specialised team with standard care. Leaving the study early was reduced (n=547, 1 RCT, RR 0.59 CI 0.4 to 0.8, NNT 9 CI 6 to 18) and compliance with treatment improved (n=507, RR stopped treatment 0.20 CI 0.1 to 0.4, NNT 9 CI 8 to 12). The mean number of days spent in hospital at one year were not significantly different (n=507, WMD, −1.39 CI −2.8 to 0.1), neither were data for ‘Not hospitalised’ by five years (n=547, RR 1.05 CI 0.90 to 1.2). There were no significant differences in numbers ‘not living independently’ by one year (n=507, RR 0.55 CI 0.3 to 1.2). At five years significantly fewer participants in the treatment group were ‘not living independently’ (n=547, RR 0.42 CI 0.21 to 0.8, NNT 19 CI 14 to 62). When phase-specific treatment (CBT) was compared with befriending no significant differences emerged in the number of participants being hospitalised over the 12 months (n=62, 1 RCT, RR 1.08 CI 0.59 to 1.99).

Phase-specific treatment E-EPA oils suggested no benefit (n=80, 1 RCT, RR no response 0.90 CI 0.6 to 1.4) as did phase-specific treatment brief intervention (n=106, 1 RCT, RR admission 0.86 CI 0.4 to 1.7). Phase-specific ACE found no benefit but participants given vocational intervention were more likely to be employed (n=41, 1 RCT, RR 0.39 CI 0.21 to 0.7, NNT 2 CI 2 to 4). Phase-specific cannabis and psychosis therapy did not show benefit (n=47, RR cannabis use 1.30 CI 0.8 to 2.2) and crisis assessment did not reduce hospitalisation (n=98, RR 0.85 CI 0.6 to 1.3). Weight was unaffected by early behavioural intervention.

Authors’ conclusions

There is emerging, but as yet inconclusive evidence, to suggest that people in the prodrome of psychosis can be helped by some interventions. There is some support for specialised early intervention services, but further trials would be desirable, and there is a question of whether gains are maintained. There is some support for phase-specific treatment focused on employment and family therapy, but again, this needs replicating with larger and longer trials.

Medical Subject Headings (MeSH): *Psychotic Disorders [diagnosis; therapy], *Schizophrenia [diagnosis; therapy], Cognitive Therapy, Early Diagnosis, Randomized Controlled Trials as Topic, Suicidal Ideation, Time Factors

MeSH check words: Humans

BACKGROUND

Schizophrenia and other functional psychoses cause enormous suffering for individuals and their families, and are a financial burden to the NHS and other health services. The estimated total cost of schizophrenia in England was £6.7 billion in 2004/05; the direct cost of treatment and care was £2 billion, whilst the indirect cost to society was £4.7 billion, and the cost of informal care and private expenditure was £615 million (Mangalore 2007). Despite new medications and the development of community care, about one-third of people with schizophrenia have a poor long-term outcome (Mason 1997). An overview of studies investigating outcomes has shown that people with schizophrenia have a one-year relapse rate of 15% to 35%, rising to 80% within five years (Larsen 1998). Achievement of full remission becomes less likely after each relapse, and about 10% of sufferers eventually commit suicide (Wiersma 1998).

Description of the condition

Schizophrenia is a chronic, relapsing mental illness and has a worldwide lifetime prevalence of about 1% irrespective of culture, social class and race. Schizophrenia is characterised by positive symptoms such as hallucinations, delusions and jumbled thinking; and negative symptoms such as apathy, poverty of speech, and withdrawal from social activities.

Description of the intervention

Early intervention in psychosis has two elements that are distinct from standard care: early detection and phase-specific treatment. Early detection may be defined as either the identification of people thought likely to develop psychosis (i.e. those who display prodromal symptoms, but have never been psychotic (Schaffner 2001)) or the identification of people who are already psychotic, but have not yet received adequate treatment (Wyatt 2001).

Phase-specific treatments are defined as treatments (psychological, social or physical) that are especially targeted at people in the prodrome or early stages of schizophrenia (Miller 1999). Phase-specific treatments may be directed at preventing progression to psychosis (in people with prodromal symptoms), or at promoting recovery (in people who have recently experienced their first episode of psychosis).

Early detection and phase-specific treatments may be provided as supplements to standard psychiatric care, or they may be provided by means of a specialised early intervention team (Garety 2000). Such teams provide care exclusively to people who have prodromal symptoms or are in early stages of schizophrenia (Edwards 2000). Prodromal patients are usually assessed by the attenuated psychotic symptom criteria, using either the criteria by Yung 2005 or the Scale of Psychotic Symptoms (SOPS, Miller 1999). A second method, is the detection of ‘basic symptoms’ developed in Germany (Schultze-Lutter 2007). When people are referred to as ‘ultra high risk’ they are using the Yung 2005 criteria. When they are referred to as early or late initial prodrome state, they are using basic symptoms.

How the intervention might work

Until recently, the orthodox approach to treating schizophrenia was to concentrate therapeutic resources on those people who developed severe and chronic disabilities (McGorry 1999). This approach has been challenged by proponents of early intervention, who have argued that greater investment of resources in the early stages of the disorder might substantially reduce the numbers of people developing chronic disabilities (Wyatt 1991). This argument has been strengthened by the observation that there may be an association between various outcome parameters and the duration of untreated psychosis (the time from the development of the first psychotic symptom to the receipt of adequate drug treatment) (Norman 2001). This has led to the proposition that untreated psychosis may be ‘toxic’ and that early intervention might prevent irreversible harm (Sheitman 1998).

Why it is important to do this review

The arguments in favour of early intervention have been so persuasive that early intervention teams are well-established in America, Europe and Australasia (Edwards 2002). In 2000, the UK government announced its intention to set up 50 early intervention teams in England to provide specialised care to all young people with a first episode of psychosis (DoH 2000). It remains unclear, however, how far these service developments are underpinned by evidence of effectiveness. There is particular concern over the ethics of early intervention with prodromal patients, when the benefits of early detection and treatment are unclear, and there is no certainty that they will go on to develop psychosis (Rosen 2000).

OBJECTIVES

To evaluate the effects of early intervention in the treatment of early psychosis.

The two specific objectives were to determine the following.

1. The effects of early detection and treatment of people with ‘prodromal’ symptoms, in terms of:

1.1 prevention of progression to full-blown psychosis;

1.2 clinical and social outcomes;

1.3 process variables and costs.

2. To determine the effects of early detection and treatment of people in their first episode of psychosis, in terms of:

2.1 clinical and social outcomes;

2.2 prevention of relapse;

2.3 process variables and costs;

2.4 reduction in duration of untreated psychosis.

We defined ‘Treatment’ as including both phase-specific treatments and care from a specialised early intervention team. We are not concerned with evaluating the accuracy of methods of predicting who is likely to develop psychosis.

METHODS

Criteria for considering studies for this review

Types of studies

We included studies if they were randomised controlled trials (RCTs). We accepted cluster-randomised trials and listed non-randomised trials in the Characteristics of excluded studies table.

In broad terms we have included two types of trial in this review.

1. Trials to prevent the development of psychosis

These studies involved treatments and/or methods of management that are given to people who are believed to be showing prodromal (pre-psychotic) symptoms and are therefore considered at high risk of developing psychosis. The primary aim of such studies was to prevent progression to psychosis, and invariably the interventions they offered were combined with some method of early detection of people at risk.

2. Trials to improve outcome in first-episode psychosis

These studies involved treatments and/or methods of management designed for people in the early stages of psychosis. The primary aim of such studies was to improve the long term outcome. Early detection might be offered in addition to the treatments, with the aim of ensuring that the treatment was offered as early as possible after the onset of psychosis.

Types of participants

1. For trials to prevent the development of psychosis, we included people who were judged by the trialists to be in a prodromal phase of psychosis, on the basis of showing prodromal symptoms (however defined).

2. For trials to improve outcome in first-episode psychosis, we included people who were in their first episode of psychosis, or were in the process of recovering from their first episode. People with psychosis were defined as those presenting with any combination of delusions, hallucinations or thought disorder, or those who had been given a diagnosis of schizophrenia or schizophrenia-like disorder, bipolar disorder (manic episode i.e. with psychotic symptoms), or depression with psychotic features.

We excluded trials where the majority of participants were suffering from a learning disability or an organic psychosis. We did not exclude anyone for reasons such as age or type of psychosis (for example, affective psychosis). Where studies included both first and second episode participants, we excluded trials if more than 10% of the participants included in the study had experienced a second episode,

Types of interventions

In trials of early intervention there are many possible combinations of intervention and control condition. This depends on: the type of participant (prodromal or first episode); whether the trial involved early detection (which could involve the whole sample or just the treatment group); the type of intervention (phase-specific or specialised team); the nature of any phase-specific treatment (cognitive therapy, family therapy etc); and the type of control (no treatment, standard psychiatric care, care from a specialised team but not phase specific intervention, etc.). In this section the most likely combinations of intervention and control conditions are listed for trials to prevent the development of psychosis and trials to improve outcome in first-episode psychosis.

1. Trials to prevent the development of psychosis in prodromal patients

These trials require prodromal patients, and since such patients do not normally present to psychiatric services, the trials therefore require some form of early detection to be applied to the whole sample. The intervention may consist of: phase-specific treatment (medication, psychological treatment or other) or care from a specialised team (which might offer phase-specific treatments). The control condition may consist of no treatment, or standard psychiatric care, or care from a specialised team (in which case the intervention will consist of care from a specialised team plus a phase specific intervention). The various types of intervention and control condition are described in more detail below.

1.1 Phase-specific treatment

In the context of preventing psychosis, phase-specific treatments are discrete interventions including medication regimes, which have been specifically developed for use in patients experiencing prodromal symptoms. A phase-specific treatment could be offered by an individual therapist or provided in the context of receiving care from a specialised team (see 1.2 below). More than one phase-specific treatment might be offered at the same time (for example, medication regime and cognitive therapy).

1.2 Care from a specialised team

In the context of preventing psychosis, this is defined as a multi-disciplinary psychiatric team, specialising in the treatment of patients with prodromal symptoms. Such a team would normally provide comprehensive psychiatric care to its patients and would be an alternative, rather than an addition, to standard psychiatric care. In the context of a trial it is likely that any specialised team would also offer phase-specific interventions.

1.3 Control conditions

In the context of preventing psychosis, the common control conditions are: no treatment; non-specific supportive therapy or care from a specialised team (which did not offer phase-specific treatments to prevent onset of psychosis).

2. Trials to improve outcome in first-episode psychosis

The intervention may consist of: early detection; phase-specific treatments (medication, psychological intervention or other) or care from a specialised team (which might offer phase-specific treatments). The control condition may consist of standard psychiatric care or care from a specialised team (in which case the intervention will consist of care from a specialised team plus a phase-specific intervention). A ‘no treatment’ control group is not an ethically acceptable option in first-episode psychosis trials. The various types of intervention and control condition are described in more detail below.

2.1 Early detection

In trials to improve outcome in first-episode psychosis it is possible to use early detection as an intervention applied to the treatment group alone; this is in contrast to the situation in trials designed to prevent psychosis (see 1. above) where early detection must be applied to both treatment and control groups. The theoretical basis for using early detection as an intervention is that shortening the duration of untreated psychosis in itself improves outcome. In trials where early detection is the intervention being tested, the unit of randomisation must be a cluster (for example, general practices or catchment areas), since it is not possible to individually randomise patients who have not yet been diagnosed.

2.2 Phase-specific treatment

In the context of improving outcome in the first episode, phase-specific treatments are discrete treatments and include medication regimes which have been specifically developed for use in the early stages of psychosis. A phase-specific treatment can be offered by an individual therapist or provided in the context of receiving care from a specialised team (see 2.3 below). More than one phase-specific treatment might be offered at the same time (for example, medication regime and cognitive therapy).

2.3 Care from a specialised team

In the context of improving outcome in first episode, this is defined as a multi-disciplinary psychiatric team, specialising in the treatment of patients with first-episode psychosis. Such a team would normally provide comprehensive psychiatric care to its patients and is an alternative, rather than an addition, to standard psychiatric care. In the context of a trial, it is likely that any specialised team would also offer phase-specific interventions.

2.4 Control conditions

In the context of improving outcome in first episode, the common control conditions are standard care, or care from a specialised team (which does not offer the phase-specific treatment being provided in the treatment arm of the trial). Standard care would be the normal service for people with severe psychiatric illness in the region where the trial took place, and would normally consist of out-patient follow up, medication, and support form a community mental health team, but would not involve any phase-specific treatment or specialised team.

3. Excluded interventions

We considered treatment with low doses of neuroleptic medication (atypical or standard) a phase-specific treatment if given to prevent progression to psychosis, or in the context of a medication protocol designed specifically for treating patients in their first episode of psychosis. However, simple comparisons of atypical neuroleptic medication versus standard neuroleptics in first-episode patients were beyond the scope of this review.

Types of outcome measures

Primary outcomes

For trials to prevent the development of psychosis (i.e. prodromal participants) the primary outcomes were as follows.

1. General

1.1 Converting to psychosis during follow-up period

For trials to improve the outcome of first-episode psychosis the outcomes were as follows.

1. General

1.1 Relapse - as defined by each study or re-admission during follow-up

Secondary outcomes

For trials to prevent the development of psychosis (i.e. prodromal participants) the secondary outcomes were as follows.

1. General

1.1 Overall functioning

1.2 Duration of hospital stay

1.3 Loss to follow up

1.4 Satisfaction with treatment - participant/carer

1.5 Remaining in contact

2. Mental state

2.1 General symptoms

2.2 Specific symptoms

2.2.1 Positive symptoms (delusions, hallucinations, disordered thinking)

2.2.2 Negative symptoms (avolition, poor self-care, blunted affect)

2.2.3 Mood - depression

3. Behaviour

3.1 General behaviour

3.2 Specific behaviours (for example, aggressive or violent behaviour)

3.2.1 Social functioning

3.2.2 Employment status during trial (employed/unemployed)

3.2.3 Occurrence of violent incidents (to self, others or property)

4. Adverse effects

4.1 General

4.2 Specific

4.2.1 Death (suicide and non-suicide)

4.2.2 Movement disorders (extrapyramidal side effects, specifically tardive dyskinesia and neuroleptic malignant syndrome)

4.2.3 Sedation

4.2.4 Dry mouth

4.2.5 Weight gain

5. Economic

5.1 Cost of care

6. Quality of life

6.1 No substantial improvement in quality of life

For trials to improve the outcome of first-episode psychosis the secondary outcomes were:

1. General

1.1 Overall functioning

1.2 Hospital readmission

1.3 Duration of hospital stay

1.4 Loss to follow-up

1.5 Satisfaction with treatment - participant/carer

1.6 Remaining in contact with services

2. Mental state

2.1 General symptoms

2.2 Specific symptoms

2.2.1 Positive symptoms (delusions, hallucinations, disordered thinking)

2.2.2 Negative symptoms (avolition, poor self-care, blunted affect)

2.2.3 Mood - depression

3. Behaviour

3.1 General behaviour

3.2 Specific behaviours (for example, aggressive or violent behaviour)

3.2.1 Social functioning

3.2.2 Employment status during trial (employed/unemployed)

3.2.3 Occurrence of violent incidents (to self, others or property)

4. Adverse effects

4.1 General

4.2 Specific

4.2.1 Death (suicide and non-suicide)

4.2.2 Movement disorders (extrapyramidal side-effects, specifically tardive dyskinesia and neuroleptic malignant syndrome)

4.2.3 Sedation

4.2.4 Dry mouth

4.2.5 Weight gain

5. Economic

5.1 Cost of care

Search methods for identification of studies

We applied no language restrictions within the limitations of the search.

Electronic searches

1. Cochrane Schizophrenia Group Trials Register (March 2009)

The register was searched using the phrase:

[early* in title, abstract or keywords of REFERENCE] or [Early* in intervention or prodromal* or early*’ in Health Care Condition of STUDY]

This register is compiled by systematic searches of major databases, hand searches and conference proceedings (see Group Module).

2. Previous searches for earlier versions of this review

Please see Appendix 1.

Searching other resources

1. Reference lists

We inspected reference lists of all identified trials and reviews for additional trials.

2. Personal contact

We contacted experts in the field within the European First Episode Network (2003) to identify unpublished trials.

Data collection and analysis

Selection of studies

We (MM and AL) searched The Cochrane Schizophrenia Group’s register. Working independently we examined the papers identified from the search strategy. We discarded obviously irrelevant publications and retained only those in which some form of early intervention had been compared against a control treatment, and obtained copies of papers relating to relevant trials. Once we had obtained these papers, we decided whether the trials were eligible. We resolved any disagreements by discussion. For the 2006 update we (MM and JR) independently inspected citations. Where disagreement occurred, we sought to resolve this by discussion, or where doubt remained, we acquired the full article for further inspection. Once we had obtained the full articles, we independently decided whether they met the review criteria. We resolved any disagreements that occurred by discussion, and when this was not possible we added trials to the list of those awaiting assessment until we acquired further information. For the 2009 update we (MM and JR) inspected all study citations identified by the searches, and obtained full reports of the studies of agreed relevance.

Data extraction and management

1. Extraction

We (MM, AL) independently extracted and entered trial data into Review Manager (RevMan) twice, cross-checking for consistency (RevMan 2008). An initial analysis included all trials meeting inclusion criteria, whilst a second sensitivity analysis excluded all but the highest quality trials (Category A and B). For the 2006 and 2010 update, we (MM and JR) independently extracted and entered data into RevMan, cross-checking again for consistency. Where disputes arose, we attempted to resolve these by discussion. When this was not possible and further information was needed to resolve the dilemma, we did not enter the data, and added this outcome of the trial to the list of those awaiting assessment.

2. Management

2.1 Forms

We extracted the data onto standard, simple forms.

2.2 Direction of graphs

Where possible, we entered data into RevMan in such a way that the area to the left of the ‘line of no effect’ indicates a ‘favourable’ outcome for early intervention. Where this was not possible, (for example, scales that calculate higher scores=improvement) we inserted a minus sign into the data tables to reverse the graphical display in RevMan analyses so that the direction of effect was clear.

2.3 Scale-derived data

Unpublished scales are known to be subject to bias in trials of treatments for schizophrenia (Marshall 2000). Therefore we only included continuous data from rating scales were if the measuring instrument had been described in a peer-reviewed journal.

2.4 Skewed data

Continuous data on outcomes in trials relevant to mental health issues are often not normally distributed. To avoid the pitfall of applying parametric tests to non-parametric data we applied the following standards to continuous final value endpoint data before inclusion: (a) standard deviations and means were reported in the paper or were obtainable from the authors; (b) when a scale started from zero, the standard deviation, when multiplied by two, should be less than the mean (otherwise the mean is unlikely to be an appropriate measure of the centre of the distribution - Altman 1996); in cases with data that are greater than the mean we entered them into the ‘Other data’ table as skewed data. Where the skewed data are derived from a trial with ≧ 200 participants, the skewed data pose less of a problem when looking at means if the sample size is large and were entered into syntheses.

If a scale starts from a positive value (such as PANSS, which can have values from 30 to 210) the calculation described above in (b) should be modified to take the scale starting point into account. In these cases skewness is present if 2SD>(S-Smin), where S is the mean score and Smin is the minimum score. We reported non-normally distributed data (skewed) in the ‘other data types’ tables. For change data (mean change from baseline on a rating scale) it is impossible to tell whether data are non-normally distributed (skewed) or not, unless individual patient data are available. After consulting the ALLSTAT electronic statistics mailing list, we entered change data in RevMan analyses and reported the finding in the text to summarise available information. In doing this, we assumed either that data were not skewed or that the analysis could cope with the unknown degree of skew.

2.5 Final endpoint value versus change data

Where both final endpoint data and change data were available for the same outcome category, only final endpoint data were presented. We acknowledge that by doing this much of the published change data may be excluded, but argue that endpoint data is more clinically relevant and that if change data were to be presented along with endpoint data, it would be given undeserved equal prominence. Where studies reported only change data we contacted authors for endpoint figures.

2.6 Common measure

To facilitate comparison between trials, we converted variables (such as days in hospital) that could be reported in different metrics (mean days per year, per week or per month) to a common metric (for example, mean days per month).

2.7 Conversion of continuous to binary

Where possible, efforts were made to convert outcome measures to dichotomous data. This could be done by identifying cut-off points on rating scales and dividing participants accordingly into ‘clinically improved’ or ‘not clinically improved’. It was generally assumed that if there had been a 50% reduction in a scale-derived score such as the Brief Psychiatric Rating Scale (BPRS, Overall 1962) or the Positive and Negative Syndrome Scale (PANSS, Kay 1986), this could be considered as a clinically significant response (Leucht 2005a; Leucht 2005b). If data based on these thresholds were not available, we used the primary cut-off presented by the original authors.

2.8 Summary of findings table

For the 2011 version of the review we had available to us the possibility of producing Summary of Findings tables. These should be considered before being biased by the results of analyses, but for us this is impossible. We have chosen to present two - but this choice is post hoc. We chose to present data from PACE-Australia and OPUS-Scandinavia as these are benchmark trials in this area and outcomes from these trials that we think to be clinically important.

Progression to psychosis
Compliance with treatment - treatment stopped in spite of need
Leaving the study early
Service use: 1. Average mean number of days per month in hospital
Service use: 2. Not hospitalised
Social outcomes: 1. Not living independently
Social outcomes: 2. Not working or in education

Assessment of risk of bias in included studies

Again working independently, we assessed risk of bias using the tool described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2008). This tool encourages consideration of how the sequence was generated, how allocation was concealed, the integrity of blinding at outcome, the completeness of outcome data, selective reporting and other biases. We would not have included studies where sequence generation was at high risk of bias or where allocation was clearly not concealed.

The categories are defined below.

- YES - low risk of bias

- NO - high risk of bias

- UNCLEAR - uncertain risk of bias

If disputes arose as to which category we should allocate a trial, again, we achieved resolution by discussion, after working with a third reviewer.

Earlier versions of this review used a different, less well-developed, means of categorising risk of bias (see Appendix 2).

Measures of treatment effect

1. Binary data

For binary outcomes we calculated an estimate of the risk ratio (RR) and its 95% (fixed-effect) confidence intervals (CI). RR is more intuitive (Boissel 1999) than odds ratios and odds ratios tend to be interpreted as RR by clinicians (Deeks 2000). This misinterpretation then leads to an overestimate of the impression of the effect. When the overall results were significant we calculated the number needed to treat/harm (NNT/NNH) using Visual Rx.

2. Continuous data

For continuous outcomes we estimated mean difference (MD) between groups. We preferred not to calculate effect size measures (standardised mean difference SMD). However, had scales of very considerable similarity been used, we would have presumed there was a small difference in measurement, and we would have calculated effect size and transformed the effect back to the units of one or more of the specific instruments.

Unit of analysis issues

1. Cluster trials

Studies increasingly employ cluster randomisation (such as randomisation by clinician or practice), but analysis and pooling of clustered data pose problems. Firstly, authors often fail to account for intra-class correlation in clustered studies, leading to a unit-of-analysis error (Divine 1992) whereby P values are spuriously low, confidence intervals unduly narrow and statistical significance overestimated. This causes Type I errors (Bland 1997; Gulliford 1999).

Where clustering had not been accounted for in primary studies, we presented the data in a table, with a (*) symbol to indicate the presence of a probable unit of analysis error. In subsequent versions of this review we will seek to contact first authors of studies to obtain intra-class correlation co-efficients (ICCs) of their clustered data and to adjust for this using accepted methods (Gulliford 1999). Where clustering has been incorporated into the analysis of primary studies, we will also present these data as if from a non-cluster randomised study, but adjusted for the clustering effect.

We have sought statistical advice and have been advised that the binary data as presented in a report should be divided by a design effect. This is calculated using the mean number of participants per cluster (M) and the ICC (Design effect=1+ (M −1)*ICC) (Donner 2002). If the ICC is not reported we assumed it to be 0.1 (Ukoumunne 1999). If cluster studies had been appropriately analysed taking into account ICCs and relevant data documented in the report, we synthesised these with other studies using the generic inverse variance technique.

2. Cross-over design

A major concern of cross-over trials is the carry-over effect. It occurs if an effect (for example, pharmacological, physiological or psychological) of the treatment in the first phase is carried over to the second phase. As a consequence, on entry to the second phase the participants can differ systematically from their initial state, despite a wash-out phase. For the same reason cross-over trials are not appropriate if the condition of interest is unstable (Elbourne 2002). As both effects are very likely in schizophrenia, we will only use data of the first phase of cross-over studies.

3. Studies with multiple treatment groups

We presented studies involving more than two treatment arms, if relevant, in comparisons.

Dealing with missing data

1. Overall loss of credibility

At some degree of loss to follow-up data must lose credibility (Xia 2007). We are forced to make a judgment where this is for the trials likely to be included in this review. Should more than 50% of data be unaccounted for by eight weeks, we did not reproduce these data or use them within analyses.

2. Intention to treat analysis

2.1 Binary data

We excluded data from studies where more than 50% of participants in any group were lost to follow-up (this did not include the outcome of ‘leaving the study early’). In studies with less than 50% dropout rate, people leaving early were considered to have had the negative outcome, For example, those lost to follow-up for the outcome of relapse were treated in the analysis as having relapsed. Suicide was treated as relapse.

2.2 Continuous data

2.2.1 Attrition

In the case where attrition for a continuous outcome is between 0% and 50% and completer-only data were reported, we have reproduced these.

2.2.2 Standard deviations

We first tried to obtain the missing values from the authors. If not available, where there were missing measures of variance for continuous data but an exact standard error and confidence interval were available for group means, and either P value or T value were available for differences in mean, we noted these, and in future versions will calculate them according to the rules described in the Handbook (Higgins 2008): When only the standard error (SE) is reported, standard deviations (SDs) can be calculated by the formula SD=SE * square root (n). Chapters 7.7.3 and 16.1.3 of the Handbook (Higgins 2008) present detailed formula for estimating SDs from P values, T or F values, confidence intervals, ranges or other statistics. If these formula do not apply, we, in the future will calculate SDs according to a validated imputation method which is based on the SDs of the other included studies (Furukawa 2006). Some of these imputation strategies can introduce error. The alternative would be to exclude a given study’s outcome and thus to lose information. We will examine the validity of the imputations in a sensitivity analysis excluding imputed values.

2.2.3 Last observation carried forward

We anticipated that in some studies the method of last observation carried forward (LOCF) would be employed within the study report. As with all methods of imputation to deal with missing data, LOCF introduces uncertainty about the reliability of the results. Therefore, where LOCF data have been used in the trial, if less than 50% of the data had been assumed, we reproduced these data and indicated that they are the product of LOCF assumptions.

Assessment of heterogeneity

1. Clinical heterogeneity

We considered all included studies initially, without seeing comparison data, to judge clinical heterogeneity. We simply inspected all studies for clearly outlying situations or people which we had not predicted would arise. When such situations or participant groups arose, we would have fully discussed these.

2. Methodological heterogeneity

We considered all included studies initially, without seeing comparison data, to judge methodological heterogeneity. We simply inspected all studies for clearly outlying methods which we had not predicted would arise. Should such methodological outliers arise we would have fully discussed these.

3. Statistical heterogeneity

3.1 Visual inspection

We visually inspected graphs to investigate the possibility of statistical heterogeneity.

3.2 Employing the I² statistic

We investigated heterogeneity between studies by considering the I² method alongside the Chi² P value. The I² provides an estimate of the percentage of inconsistency thought to be due to chance (Higgins 2003). The importance of the observed value of I² depends on i. magnitude and direction of effects and ii. strength of evidence for heterogeneity (e.g. P value from Chi² test, or a confidence interval for I²). We interpreted I² estimate greater than or equal to 50% accompanied by a statistically significant Chi² statistic as evidence of substantial levels of heterogeneity (Section 9.5.2 - Higgins 2008). When substantial levels of heterogeneity were found in the primary outcome, we explored reasons for heterogeneity (Subgroup analysis and investigation of heterogeneity).

Assessment of reporting biases

Reporting biases arise when the dissemination of research findings is influenced by the nature and direction of results. These are described in section 10.1 of the Handbook (Higgins 2006). We are aware that funnel plots may be useful in investigating reporting biases but are of limited power to detect small-study effects (Egger 1997). We did not use funnel plots for outcomes where there were 10 or fewer studies, or where all studies were of similar sizes. In other cases, where funnel plots were possible, we sought statistical advice in their interpretation.

Data synthesis

Where possible we employed a fixed-effect model for analyses. We understand that there is no closed argument for preference for use of fixed-effect or random-effects models. The random-effects method incorporates an assumption that the different studies are estimating different, yet related, intervention effects. This does seem true to us; however, random-effects does put added weight onto the smaller of the studies - those trials that are most vulnerable to bias. For this reason we favour using the fixed-effect model.

Subgroup analysis and investigation of heterogeneity

1. Subgroup analyses

We did not anticipate subgroup analyses.

2. Investigation of heterogeneity

If inconsistency was high, we have reported this. First we investigated whether data had been entered correctly. Second, if data had been correct, we visually inspected the graph and successively removed studies outside of the company of the rest to see if homogeneity was restored. Should this occur with no more than 10% of the data being excluded, we have presented data. If not, we have not pooled data and have discussed relevant issues.

Should unanticipated clinical or methodological heterogeneity be obvious we simply stated hypotheses regarding these for future reviews or versions of this review. We did not anticipate undertaking analyses relating to these.

Sensitivity analysis

For the 2011 version of this review we did not anticipate undertaking any additional sensitivity analyses.

RESULTS

Description of studies

See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification; Characteristics of ongoing studies.

For substantive descriptions of studies, please see Characteristics of included studies; Characteristics of excluded studies; Characteristics of ongoing studies.

Results of the search

The original search strategy identified 9279 abstracts, of which 184 referred to potentially eligible studies and 155 reviews of early intervention (Figure 1). From these we identified 100 relevant studies, of which 43 did not meet inclusion criteria. We were able to include three studies and the remainder are awaiting classification. For substantive descriptions of studies, please see the Characteristics of included studies and Characteristics of excluded studies tables.

For the 2006 update search we identified 159 new citations and were able to include four additional studies.

During the 2009 update search we identified 830 references (from 420 studies) and were able to include 11 additional studies (Alvarez-Spain; Amminger-Austria; Berger-Australia; Edwards-Australia; EIPS-Germany; Jackson-Australia; Killackey-Australia; Leavey-UK; LEO-CAT-UK; LIPS-Germany; Uzenoff-USA).

Included studies

We included 18 studies with 1808 participants.

1. Methods

For description of methods - please see Risk of bias in included studies.

2. Participants and setting

2.1 Participants with prodromal symptoms - and setting

Six trials (Amminger-Austria, EIPS-Germany, EDIE-UK, LIPS-Germany, PACE-Australia, PRIME-USA) were concerned with preventing the onset of psychosis.

EIPS-Germany undertook the research in community settings in Cologne, Bonn, Dusseldorf, and Munich. The participants had a mean age of 26 years and were judged at risk of psychosis because they met the criteria for the Early Initial Prodrome State according to the presence of Basic Symptoms. The LIPS-Germany study was undertaken in the same settings as EIPS-Germany, but included late prodrome participants who were at risk of psychosis according to the Basic Symptom criteria for Late Initial Prodrome State (where conversion to psychosis is considered more imminent than in the Early Initial Prodrome State). Participants had a mean age of 25 years. EDIE-UK recruited participants from primary care teams (general practitioners, practice nurses and psychological therapists), student counselling services, accident and emergency departments, specialist services (community drug and alcohol teams, child and adolescent psychiatry and adult psychiatry services) and voluntary sector agencies. Participants had a mean age of 21 years and were judged to have an ‘ultra high risk’ of developing a first episode of psychosis (Yung’s Criteria Yung 2005).PACE-Australia recruited participants referred to the Personal Assessment and Crisis Evaluation clinic, which is part of the EPPIC programme. Participants were from 14 to 30 years of age, and met Yung criteria for an ‘ultra high risk’ mental state (see included studies table for details). PRIME-USA recruited people from referrals and by participants responding to study advertisements. Participants were aged from 12 to 36 years with a diagnosis of being at risk of developing psychosis according to SOPS criteria (similar to Yung’s criteria for the ‘ultra high risk’ mental state). Participants were recruited at four sites (three in the USA and one in Canada). Amminger-Austria recruited outpatients at the Vienna General Hospital who were at risk of developing first-episode psychosis according to Yung’s criteria for the ‘ultra high risk’ mental state. Participants were eligible if they were aged from 13-24 years; the mean age of the participants was 16.4 years.

2.2 Participants with first-episode psychosis and setting

Twelve trials were concerned with improving outcome in first-episode psychosis (Alvarez-Spain; Berger-Australia; Edwards-Australia; Jackson-Australia; Killackey-Australia; Leavey-UK; LEO-CAT-UK; LifeSPAN-Australia; Linszen-Amsterdam; OPUS-Scandinavia; Uzenoff-USA; Zhang-China).

Alvarez-Spain included drug naïve participants with a DSM-IV diagnosis of psychosis, and a mean age of 26 years; the trial was set in the community. Edwards-Australia included participants with first-episode psychosis diagnosed as having a psychotic disorder using DSM-IV criteria. The study was undertaken at the EPPIC centre in Melbourne, Australia. Berger-Australia recruited participants in Melbourne, Australia (EPPIC Centre); participants had an average age of 20 years, and a mean antipsychotic treatment exposure prior to study of 17.5 days. Jackson-Australia included people with a mean onset of psychosis at 22 years; the settings used were at the participant’s home, a neutral location or the EPPIC centre. Leavey-UK included first-episode psychosis patients who had been diagnosed within the last six months and were recruited from psychiatric services in North London. LEO-CAT-UK included participants with first-episode psychosis (Yung’s criteria) with a mean age of 23 years. The study was undertaken in community settings within the borough of Lambeth, London, UK. LifeSPAN-Australia recruited participants from the Western region of Melbourne, Australia and is part of the EPPIC programme, which includes an early detection and crisis assessment team. Participants were aged 15 to 29 years, and were acutely suicidal. Linszen-Amsterdam recruited participants aged from 15 to 26 who were experiencing their first episode of schizophrenia and living in close contact with parents or relatives. All participants were recruited from an adolescent clinic and had to agree to an initial three months’ inpatient programme before randomisation. Subsequent treatment took place on an outpatient basis. OPUS-Scandinavia recruited first-episode psychosis (ICD 10) patients from inpatient and outpatient departments in Denmark; participants were aged 18 to 45. Zhang-China recruited only men who had just been discharged from Suhoz Psychiatric Hospital in China, following their first admission for schizophrenia. The intervention and standard care were provided on an outpatient basis. Killackey-Australia enrolled participants from the EPPIC programme with first-episodepsychosis; participants had a mean age of 21 years. Uzenoff-USA recruited participants with first episode schizophrenia in the USA.

3. Study size

OPUS-Scandinavia was the largest study, and had a sample size (n=547) which was arrived at using a pre-study power calculation. The other trials were small and, in ascending order of size, were:Uzenoff-USA (24), Killackey-Australia (41), Edwards-Australia (47), LifeSPAN-Australia (56), PACE-Australia (59), EDIE-UK (60), PRIME-USA (60), Alvarez-Spain (61), Jackson-Australia (62), Linszen-Amsterdam (76), Berger-Australia (80), Amminger-Austria (81), Zhang-China (83), Leavey-UK (106), LEO-CATUK (113), LIPS-Germany (124) and EIPS-Germany (128).

4. Intervention

4.1 Trials to prevent the onset of psychosis

Six trials (Amminger-Austria; EIPS-Germany; LIPS-Germany; EDIE-UK; PACE-Australia; PRIME-USA) were concerned with preventing the onset of psychosis.

Amminger-Austria compared omega-3 fatty acids (eicosapentaenoic acid and docosahexaenoic acid) with placebo over three months in adolescents at risk of first-episode psychosis according to the criteria of Yung 2005.

EIPS-Germany employed a complex intervention consisting of 12 months of cognitive behaviour therapy (CBT) delivered in group and individual therapy sessions, supplemented by cognitive remediation therapy. The participants were at risk of developing first episode of psychosis and met the early initial prodromal state criteria. The control group were given supportive therapy sessions which only provided minimal support involving psychoeducation and counselling.

LIPS-Germany employed a needs focused intervention combined with amisulpride (mean dose 118 mg/day) in participants judged at risk of psychosis because of the presence of prodromal symptoms. The control group also received the needs focused intervention but without amisulpride. The needs focused intervention included psychoeducation, crisis intervention, family counselling and assistance with education or work related difficulties according to the patient’s need.

The Early Detection and Intervention Evaluation trial (EDIEUK) used cognitive therapy, limited to a maximum of 26 sessions over six months, following the principles developed by Beck 1976. The therapy was problem-orientated and time limited and was carried out by experienced cognitive therapists. Both control and treatment group received regular monitoring. Whilst participants (treatment and control) were not given medication, both treatment and control received elements of case management in order to resolve crises regarding social issues and mental health risk.

In PACE-Australia the intervention involved prescription of low dose risperidone (1-2 mg/day) combined with modified CBT, which aimed to enhance understanding and control of symptoms. Both the intervention and control groups also received case management from a PACE therapist. This involved supportive psychotherapy, assistance with accommodation and education/employment, and family support. Participants in the control and intervention groups received standard treatment if they developed psychosis, but control patients were not otherwise prescribed neuroleptics. Both groups could be prescribed antidepressants and benzodiazepines.

The Prevention through Risk Identification Management and Education study (PRIME-USA) randomised participants to olanzapine 5-15 mg/day (mean 8 mg/day) or placebo for one year, and then followed up for a further year without medication. Individual and family psychosocial interventions with supportive and psychoeducational components were available to all patients during the first year. The nature of the psychoeducational components varied across sites despite efforts to apply them in a uniform way. The psychosocial intervention available at the New Haven centre was modelled on the Problem Solving Training approach (D’Zurilla 1971; D’Zurilla 1986).

4.2 Trials to improve the outcome of first-episode psychosis

In Alvarez-Spain, drug naive first-episode participants were randomised to three different antipsychotics (risperidone, olanzapine, haloperidol) and then randomised to either Early Behavoural Intervention or the control group (routine clinical care).

In Berger-Australia, the first-episode psychosis participants were given ethyl-eicosapentaenoic acid oil (E-EPA) at a dose of 500 mg twice Daily, with a flexible dose of atypical antipsychotics. The control group received placebo capsules with a flexible dose of atypical antipsychotics.

In Edwards-Australia, the intervention group received a behavioural modification intervention, Cannabis and Psychosis Therapy (CAP). This consisted of weekly sessions of CBT provided by trained clinicians over three months. The aim of the CAP intervention was to reduce cannabis intake and to improve clinical and psychosocial functioning. CAP involved an assessment of engagement, followed by education about cannabis and psychosis and developing motivation to change. The focus of therapy was determined by the phase of commitment to change and could include further educational sessions, motivational interviewing, goal setting, and discussion about relapse prevention. An active control group was used which consisted of psychoeducation, which explained psychosis, medication and other treatments, and relapse prevention, but did not discuss cannabis.

In Jackson-Australia, the intervention group received CBT with 20 sessions provided for 45 minutes, plus antipsychotics. The control group were given a befriending service in addition to antipsychotics.

In Killackey-Australia, the intervention group received individual placement and support, which is an intervention designed to help people with mental illness to find and keep competitive employment. The support provided in the programme continued after employment was obtained, and was adapted to the needs of the individual. The control group received treatment as usual.

In Leavey-UK, the intervention group received a brief intervention and treatment as usual. The brief intervention was provided over seven sessions, lasting about one hour, and included: information gathering from the relative, plus sessions on: psychotic illness, symptoms and early warning signs, treatment, help seeking; coping strategies, problem solving and communication with the patient. The control group were given treatment as usual.

LEO-CAT-UK was a cluster-randomised trial in which primary care (GP) practices were randomly allocated to receive training in early detection of psychosis and direct access to LEO-CAT (a specialised treatment team for first-episode psychosis). The control group of General Practice clinics did not receive training in early detection and continued to refer new cases of psychosis to local mental health services who could then refer on to the LEO-CAT programme.

In LifeSPAN-Australia, the intervention group received standard clinical care plus LifeSPAN therapy which draws on the experience at EPPIC with Cognitive Orientated Therapy for Early Psychosis (COPE) and suicide manuals such as Choosing to Live and Cognitive Therapy of Suicide Behaviour. Four phases are used for the intervention: (a) initial engagement, (b) suicide risk assessment/formulation, (c) cognitive modules and (d) final closure/handover.

In Linszen-Amsterdam the intervention was behavioural family therapy for one year. Eighteen family therapy sessions were held over a 12-month period. Each family was treated by two co-therapists, from a team of two psychologists and one social worker, all of whom had at least one year of experience in providing family interventions for schizophrenia. The intervention was based on the behavioural family management approach of Falloon 1984 and involved psychoeducation, communication training and development of problem solving skills. Both intervention and control groups also received care from a specialised first-episode team involving individual-oriented therapy consisting of maintenance medication and disease and stress management.

In OPUS-Scandinavia, participants received integrated treatment or standard care. Integrated treatment consisted of high fidelity assertive community treatment supplemented by behavioral family therapy and social skills training. Standard care consisted of care at a community mental health centre. All participants were offered antipsychotic drugs according to guidelines from the Danish Psychiatric Society, which recommends a low-dose atypical antipsy chotic strategy for first episodes of psychotic illness. Each participant was usually in contact with a physician, community mental health nurse and in some cases a social worker. In a small proportion of cases, standard care also included psychosocial interventions such as training in social skills or Daily living activities, or supportive contacts with the family. Antipsychotics were given to both groups based on the psychiatrists’ clinical assessment.

The Uzenoff-USA study provided participants with Adherence Coping Education (ACE) which consists of 14 sessions lasting between 30 and 45 minutes over six months. It is a manual-based psychotherapy grouped into four phases: (1) establishing therapeutic alliance; (2) promoting treatment adherence; (3) developing a plan for maintenance treatment and (4) rehabilitation. The control group received supportive therapy which involved (1) establishing the therapeutic relationship, and (2) providing emotional support plus discussion of non-illness issues or topics.

Zhang-China also used family therapy, but in the form of group and individual family sessions which were delivered on an outpatient basis over the 18-month follow-up period. Both intervention and control groups also received care from the outpatients department, (consisting of medication and review) but no regular appointments or community follow-ups were provided.

5. Trial duration

The six trials concerned with preventing the onset of psychosis reported data from between two months and two years:EIPS-Germany at 12 months; EDIE-UK at 12 months and 36 months; PACE-Australia at six and 12 months (the first six months being the period during which the intervention was received);PRIME-USA at two months, 12 months (first 12 months study intervention given) and 24 months (last 12 months without intervention); LIPS-Germany at three months (although the study is planned to last 24 months), and Amminger-Austria at three months.

The 12 trials concerned with improving outcome of first-episode psychosis also reported data at various time points (range three months to five years): Alvarez-Spain reported data at 13 weeks; Berger-Australia reported data at three months; Jackson-Australia followed participants for 12 months; Leavey-UK reported data at nine months. Four trials (Edwards-Australia; Killackey-Australia; LifeSPAN-Australia; Uzenoff-USA) reported data at six months. Linszen-Amsterdam reported at 12 months following an initial three-month inpatient admission and also reported a five-year follow-up, but data were provided for the whole sample only, not by group allocation. OPUS-Scandinavia reported data at 12 and 24 months, whereas Zhang-China reported at 18 months.

6. Outcomes

6.1 Non-scale data

We were able to report dichotomous data on suicide, death, leaving the study early, conversion to psychosis, adverse effects, hospital admission, days in hospital, compliance with medication, antipsychotic drug use, living independently and employment.

6.2 Scale derived data

Only details of the outcome scales that provided usable data are shown below. Reasons for exclusions of data are given under ‘Outcomes’ in Characteristics of included studies.

6.2.1 Global state scales

a. Global Assessment of Functioning - GAF (APA 1994)

This is an observer rated scale for measuring overall severity of functional impairment. GAF consists of nine behavioural descriptors. Patients are rated between 0 (most severe) and 90 (least severe) for each descriptor. PRIME-USA, PACE-Australia, OPUS-Scandinavia and LIPS-Germany reported data from this scale.

b. Clinical Global Impression - CGI (Guy 1970)

The CGI is a three-item scale commonly used in studies on schizophrenia that enables clinicians to quantify severity of illness and overall clinical improvement. The items are: severity of illness; global improvement and efficacy index. A seven-point scoring system is usually used with low scores indicating decreased severity and/or greater recovery. PRIME-USA reported data from this scale.

c. Knowledge About Psychosis Questionnaire - KAPQ (Birchwood 1992)

This questionnaire tests the patients understanding about psychosis and treatments. Data from this scale were reported by Edwards-Australia.

6.2.2 Mental state scales

a. Brief Psychopathological Rating Scale - BPRS (Overall 1962)

The BPRS is an 18-item scale measuring positive symptoms, general psychopathology and affective symptoms. The original scale has 16 items, but a revised 18-item scale is commonly used. Scores can range from 0-126. Each item is rated on a seven-point scale varying from ‘not present’ to ‘extremely severe’, with high scores indicating more severe symptoms. Data from this scale were reported by Edwards-Australia. In PACE-Australia the scale was used primarily to report severity of psychotic symptoms.

b. Positive and Negative Symptom Scale - PANSS (Kay 1987)

The Positive and Negative Symptom Scale was developed from the BPRS and the Psychopathology Rating Scale. It is used as a method for evaluating positive, negative and other symptom dimensions in schizophrenia. The scale has 30 items, and each item can be defined on a seven-point scoring system varying from one (absent) to seven (extreme). This scale can be divided into three sub-scales for measuring the severity of general psychopathology, positive symptoms (PANSS-P) and negative symptoms (PANSS-N). A low score indicates low levels of symptoms. EDIE-UK used this scale to determine transition to psychosis. PRIME-USA and LIPS-Germany reported data from the PANSS.

c. Scale of Psychotic Symptoms - SOPS (Miller 1999)

The SOPS scale was modelled on the PANSS scale and is designed to measure the presence/absence of prodromal states. It consists of five positive symptom items, six negative symptom items, four disorganisation symptoms items, and four general symptom items. Each has a severity rating from 0 (never, absent) to six (severe/extreme - and psychotic for the positive items). The severity of the prodromal state is based on the sum of the rating from the SOPS items and ranges between 0 and 114. PRIME-USA reported data from this scale.

d. Hamilton Rating Scale for Anxiety - HRSA (Hamilton 1959)

The Hamilton Anxiety Scale (HAMA) is a rating scale developed to quantify the severity of anxiety symptoms, often used in psychotropic drug evaluation. It consists of 14 items, each defined by a series of symptoms. Each item is rated on a five-point scale, ranging from 0 (not present) to 4 (severe). The 14 items consist of: anxious mood; tension; fears; insomnia; intellectual; depressed mood; somatic complaints (muscular); somatic complaints (sensory); cardiovascular symptoms; respiratory symptoms; gastrointestinal symptoms; genitourinary symptoms; autonomic symptoms and behaviour at Interview. Higher scores indicate greater anxiety. PACE-Australia reported data from this scale.

e. Hamilton Rating Scale for Depression - HRSD (Hamilton 1960)

This is an interviewer rated scale for measuring depression. It is used for quantifying the results of an interview and depends on the skill of the interviewer in eliciting the necessary information. It contains 17 variables measured on either a five point or a three-point rating scale. The variables include: depressed mood; suicide; employment and loss of interest; retardation; agitation; gastrointestinal symptoms; general somatic symptoms; hypochondriasis; loss of insight and loss of weight. Higher scores indicate more severe depression. PACE-Australia reported data from this scale.

f. Calgary Depression Rating Scale - CDRS (Addington 1990)

The Calgary Depression Scale for Schizophrenia is a nine-item scale ( 0=absent; 1=mild; 2=moderate; 3=severe.) that was specifically developed for assessment of depression in patients with schizophrenia. It has been evaluated in both relapsed and remitted patients, and is provided as a semi-structured interview. High scores indicated worse outcome. Uzenoff-USA reported data from this scale.

g. Montgomery Asberg Depression Rating Scale - MADRS (Montgomery 1979)

This is a 65-item comprehensive psychopathology scale used to identify the 17 most commonly occurring symptoms in primary depressive illness. Ratings are based on 10 items, with higher scores indicating more symptoms. This scale was used by LIPS-Germany.

h. Beck Depression Inventory - BDI-SF (Beck 1961)

This is a 21-item self-rating scale for depression. Each item comprises four statements (rated 0-4) describing increasing severity of the abnormality concerned. The person completing the scale is required to read each group of statements and identify the one that best describes the way they have felt over the preceding week. A total of 12/13 is an indicative score for presence of significant depression. The short form of this scale was used by Edwards-Australia.

i. Presence of Psychosis Scale - POPS (Olsen 2006)

The Presence of Psychosis Scale (POPS), is part of the Structured Interview for Prodromal Syndromes scale (SIPS). It marks onset of psychosis by the presence of positive symptoms at the psychotic level of intensity and of sufficient frequency and duration. PRIME-USA reported data from this scale.

j. Scale for the Assessment of Negative Symptoms - SANS (Andreasen 1983)

This is also an interviewer rated scale for measuring the severity of negative symptoms of schizophrenia such as alogia, affective blunting, avolition-apathy, anhedonia-asociality and attention impairment. Items are rated on a six-point scale with higher scores indicating more symptoms. Edwards-Australia and PACE-Australia reported data from this scale.

k. Young Mania Scale - YMS (Young 1978)

Again an interviewer rated scale, but this time for measuring the severity of symptoms of mania. Higher scores indicate more severe symptoms. PRIME-USA and PACE-Australia reported data from this scale.

6.2.3 Social functioning

a. Social Functioning Scale II - SAS II (Weissman 1976)

The SAS II is an interviewer-administered scale adapted from the self-report Social Adjustment Scale for use with people with schizophrenia. It contains 52 questions that are administered in a semi-structured interview by a trained rater. The SAS II assesses current functioning with scores ranging from 1 to 5, with higher scores indicating worse functioning. EIPS-Germany reported data from this scale.

b. Social and Occupational Functioning Assessment Scale - SOFAS (APA 1994)

The SOFAS is a new instrument similar to the Global Assessment of Functioning is format, and attempts to assess the social and or occupational functioning independent of the overall severity of the illness. Higher scores indicate worse social functioning. Edwards-Australia and Jackson-Australia reported data from this scale.

6.2.4 Adverse effects

a. Simpson Angus Scale - SAS (Simpson 1970)

The SAS is a 10-item scale used to evaluate the presence and severity of drug-induced parkinsonian symptoms. The 10 items focus on rigidity rather than bradykinesia and do not assess subjective rigidity or slowness. The scale comprises a 10-item rating scale, each item rated on a five-point scale with zero meaning the complete absence of condition and four meaning the presence of condition in extreme. A low score indicates low levels of parkinsonism. PRIME-USA reported data from this scale,

b. Barnes Akathisia Rating Scale - BAS (Barnes 1989)

The Barnes Akathisia Rating Scale is a four-item scale to assess the presence and severity of drug-induced movement disorder akathisia. It is a widely used comprehensive rating scale for akathisia. Items include restless movements that characterise akathisia, the subjective awareness of restlessness and any distress associated with the condition. These items are rated from zero (normal) to three (severe). In addition, there is an item for rating the global severity that starts from zero (absent) to five (severe). A low score indicates low levels of akathisia. PRIME-USA reported data from this scale.

c. Abnormal Involuntary Movement Scale - AIMS (Guy 1976)

The Abnormal Involuntary Movement Scale has been used to assess abnormal involuntary movements associated with antipsychotic drugs, such as tardive dyskinesia and chronic akathisia, as well as ‘spontaneous’ motor disturbance related to the illness itself. Tardive dyskinesia is a long-term, drug-induced movement disorder. However, using this scale in short-term trials may also be helpful to assess some rapidly occurring abnormal movement disorders such as tremor. Scoring consists of rating movement severity in the anatomical areas (facial/oral, extremities, and trunk) on a five-point scale (0-4). A low score indicates low levels of dyskinetic movements. PRIME-USA reported data from this scale.

6.2.5 Quality of life

a. Quality of Life Scale - QLS (Heinrichs 1984)

This is a semi-structured interview administered and rated by trained clinicians. It contains 21 items rated on a seven-point scale based on the interviewer’s judgement of patient functioning. Higher scores indicate better quality of life. PACE-Australia and Uzenoff-USA reported data from this scale.

6.2.6 Satisfaction with care

a. The Client Satisfaction Questionnaire - CSQ-8 (De-Wilde 2005)

The CSQ-8 is an eight-item self-report of global measure of patient satisfaction with services. The CSQ is substantially correlated with treatment dropout, number of therapy sessions attended, and with change in client-reported symptoms. The CSQ-8 consists of eight items rated on a four-point Likert scale. The items are concerned with quality of services received, how well services met the client’s needs and general satisfaction. The total score ranges from eight to 32. Higher scores indicate greater satisfaction of the responders OPUS-Scandinavia reported data from this scale.

6.2.7 Substance use

a. Cannabis and Substance Use Assessment Schedule - CASUAS (Wing 1990)

This scale measures the percentage of days using cannabis in the past four weeks and includes an index of severity of cannabis use. The scale is modified from the Schedule for Clinical Assessment on Neuropsychiatry and includes similar information to the Addiction Severity Index. Data from this scale were used by Edwards-Australia.

6.3 Redundant data

Some studies reported data only as P values or statements of significant or non-significant differences, and other continuous data could not be extracted because the number of participants was missing or standard deviations were not reported.

Excluded studies

There are currently 68 excluded studies. We have summarised reasons for exclusion in Table 1.

Table 1.

Reasons for excluding studies

Totals	Reasons	Totals	Reasons	Totals	References
68	Randomised	32	not first episode	13	Craig 2004b, Drury 2000, Grawe 1998, Jenner 2004, Jones 2005, Jolley 2003, Keshavan 1998, Kuipers 2004, Li 2004, Nuechterlein 2005, Power 2004, SOCRATESUK, Ueland 2004, Wykes 2007
	Randomised	32	drug studies	16	Anonymous 1987, Crow 1986, Davidson 2004, Emsley 2004, Emsley 1999, Gaebel 2004, Heydebrand 2004, Keefe 2000, Kopala 2003, Lieberman 2005b, Sanger 1999, Perez 2003, Purdon 2000, Schooler 2003, Wang 2000, Wunderink 2003
	Not randomised	36	descriptions of services	7	Bao 2005, Birchwood 1989, Clare 1994, Fisher 2001, Parlato 1999, Welch 2000, Whitwell 2000, Wieneke 2000
			before and after studies	12	Addington 1999, Alanen 1994, Albiston 1998, Culberg 1998, DeHaan 1997, Fitzgerald 1998, Fresan 2001, Newton 2005, Rund 1994, Szymanski 1994, Whitehorn 1998, Yap 2001
			quasi-experimental studies	3	COPE-Melbourne, TIPS 2006, Walczewski 1998
			inadequate randomisation	2	McCay 2007, Mosher 1975
			uncontrolled studies	13	Agius 2007, Falloon 1992, Hartmann 1974, Jenner 2001, Kadota 1992, Kauranen 2000, Keshavan 1998, Malla 2001, McGorry 1996, Mottaghipour 2000, Thomas 1979, Turetz 1997, Zhang-Wong 1999

Study	Duration of follow up (months)
Study	6	12	18	24
Amminger-Austria	No data	No data	No data	No data
Berger-Australia		86%
EDIE-UK		60%
Edwards-Australia	No data	No data	No data	No data
EIPS-Germany	No data	No data	No data	No data
Jackson-Australia		82%
Killackey-Australia	75%
Leavey-UK	79% (9 months)
LEO-CAT-UK	No data	No data	No data	No data
LifeSPAN-Australia	75%
Linszen-Amsterdam*		unclear
LIPS-Germany				61%
OPUS-Scandinavia		77%		67%
PACE-Australia		100%
PRIME-USA		84%		72%
Uzenoff-USA	79%
Zhang-China			94%

Methods	Allocation: randomised (by computer random numbers into 4 blocks). Blinding: single blind. Setting: community. Duration: interventions offered for 3 months.
Participants	Diagnosis: drug-naive first-episode psychosis (DSM-IV). N=61. Sex: 46 M, 15 F. Age: mean 26 years. Inclusion criteria: 15 to 60 years, DSM criteria for schizophrenia, schizophreniform disorder, schizoaffective disorder, delusional disorder, brief reactive psychosis, or psychosis not otherwise specified. Excluded: patients with neurological disease, head injury, mental retardation, and drug dependence. History: all patients experiencing their first episode of psychosis and had not received more than 6 weeks of adequate antipsychotic treatment
Interventions	1. Early Behavoural Intervention. N=28. 2. Routine Care Intervention*. N=33. Interventions started when participants reached a minimum clinical stabilisation, defined by a global score of < 5 on the SAPS scale, or otherwise treatment began within 6 weeks of randomisation
Outcomes	Body weight. Body Mass Index. Unable to use - Leaving the study early- no data. Mental state: SAPS (no data reported).
Notes	* This package of care was specifically designed to teach strategies to enhance control over factors associated with antipsychotic-induced weight gain and consisted of 8 flexible intervention modules that incorporated behavioural interventions, nutrition and exercise. ** The control group were informed about potential weight gain and advised to increase their exercise and limit food intake. Participants were first randomised to 3 antipsychotics (olanzapine mean dose 13 mg/day, risperidone mean dose 4.2 mg/day or haloperidol mean dose 4.9 mg/day), then randomly assigned to either EBI or RCI
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Low risk	Randomised using computer generated random numbers
Allocation concealment?	Unclear risk	No details.
Blinding? All outcomes	High risk	Single
Incomplete outcome data addressed? All outcomes	High risk	Study attrition not reported
Free of selective reporting?	High risk	SAPS data not reported
Free of other bias?	Unclear risk	No details.

Methods	Allocation: randomised, no further details. Blinding: double blind. Setting: Vienna, Austria. Inclusion criteria: not stated. Exclusion criteria: not stated. Follow up: 12 weeks.
Participants	Diagnosis: Adolescents at risk of 1st episode psychosis (Yung criteria). N=81*. Age: range 3-24 years; mean age 16.4. Sex: no details. History: no details.
Interventions	1. Omega-3 fatty acids: dose eicosapentaenoic acid 0.84 g/day; docosahexaenoic acid 0.7 g/day. N=38. 2. Placebo. N=38.
Outcomes	Transition to psychosis**. Unable to use - Leaving the study early Mental state: BPRS, PANSS (no usable data) Global state: GAF (no usable data). Adverse effects: UKU.
Notes	5 participants not accounted for. *Operationally defined, based on Yung et al’s criteria, using cut-off points on PANSS subscales, (4 or more on hallucinations, 4 or more on delusions, and 5 or more on conceptual disorganization), and the frequency of symptoms (at least several times a week) and their duration (more than one week)
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Unclear risk	Randomised, no further details.
Allocation concealment?	Unclear risk	No described.
Blinding? All outcomes	Unclear risk	Double blind, untested.
Incomplete outcome data addressed? All outcomes	High risk	Study attrition not described.
Free of selective reporting?	Unclear risk	No details.
Free of other bias?	Unclear risk	No details.

Methods	Allocation: randomised. Blinding: double blind. Setting: single centre Melbourne, Australia (EPPIC Centre). Inclusion criteria: age between 15-29 years, currently psychotic. Exclusion criteria: drug induced psychosis; first-episode mania; psychotic disorders due to organic illness. Follow-up: 12 weeks.
Participants	Diagnosis: psychosis (DSM-IV). N=80. Age: mean 20 years. Sex: male and female. History: first-episode psychosis; mean antipsychotic treatment prior to study 17.5 days
Interventions	1. Ethyl-Eicosapentaenoic Acid oil (E-EPA): dose 500 mg/bid, plus flexible dosage of atypical antipsychotics. N=40. 2. Placebo capsules: plus flexible dosage of atypical antipsychotics. N=40. Benzodiazepines, chlorpromazine or zuclopenthixol acetate allowed for behavioural control when indicated
Outcomes	Leaving the study early. Global state: Not responded to treatment. Unable to use - Mental state: BPRS, SANS, MADRAS (no data). Global state: CGI, GAF (no data). Social functioning: SOFAS no data). Adverse events: SAS, UKU (no usable data). Social functioning: SAS II.
Notes
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Unclear risk	Randomised, no further details.
Allocation concealment?	Unclear risk	No details.
Blinding? All outcomes	Unclear risk	Double blind, untested.
Incomplete outcome data addressed? All outcomes	Low risk	Study attrition reported.
Free of selective reporting?	Unclear risk	No details.
Free of other bias?	Low risk	Funded by Swiss National Science Foundation; Margaret and Walter Lichtenstein Foundation, National Health and Medical Research Council of Australia’ the Colonial Foundation of Australia, and a National Alliance for Reasearch on Schizophrenia and Depression

Methods	Allocation: randomised, stratified according to gender and genetic risk (independent clerical worker, sealed envelopes). Blinding: single blind (raters), attempts made to keep assessors blind*. Setting: community, Salford, Manchester. Inclusion criteria: based on an adaptation of the PACE criteria, and an age range between 16-36. Exclusion criteria: current of past receipt of antipsychotic medication. Follow-up: 1 year and 3 years. Evaluation: conducted by research assistants.
Participants	Diagnosis: ultra high risk of developing 1st episode of psychosis (Yung modified criteria). N=60. Age: mean 21 years. Sex: male and female. History: not reported.
Interventions	1. Cognitive therapy: dose maximum of 26 sessions, over 6 months. N=37 2. Monitoring control group. N=23.
Outcomes	Leaving the study early. Transition to psychosis (based on PANSS criteria). Unable to use - Transition to psychosis (3 year data, 53% lost to follow-up). Mental state: PANSS (no usable data). Global state: GAF, GHQ (no usable data). Sociotropy - Autonomy Scale (no usable data). Meta-Cognitions Questionnaire (no usable data). Oxford-Liverpool Inventory of Feelings and Experiences (no usable data)
Notes	*Blinding was not adequately maintained due to participants divulging information about their therapist, or used language that suggested they were receiving cognitive therapy. Three-year outcome data not used due to > 50% study attrition
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Unclear risk	Randomised, no further details.
Allocation concealment?	Low risk	By independent clerical worker, using sealed envelopes.
Blinding? All outcomes	High risk	Single blind.
Incomplete outcome data addressed? All outcomes	Low risk	Study attrition reported.
Free of selective reporting?	Unclear risk	No details.
Free of other bias?	Low risk	Funded by North-West NHS Executive.

Methods	Allocation: randomised, computer generated, placed in sealed envelopes. Blinding: single; attempts to maintain rater blindness included use of separate rooms and admin procedures for staff. Duration: 3 months intervention phase followed by 6 months of follow-up. Country: Melbourne, Australia. Setting: youth mental health service; early psychosis, prevention and intervention centre (EPPIC). Analyses: ITT analyses were used and last observation carried forward
Participants	Diagnosis: first-episode psychosis (DSM-IV). N=47. Age: 15-29 years. Sex: male and female. History: patients continuing to use cannabis after initial treatment for first episode psychosis. Inclusion criteria: DSM-IV diagnosis of a psychotic disorder (i.e. schizophrenia, schizophreniform, schizoaffective, delusional disorder, bipolar disorder, major depressive disorder with psychotic features, psychosis not otherwise stated, brief reactive psychosis). Exclusion criteria: only participants with at least 10 weeks continuous cannabis usage prior to study were eligible for study inclusion
Interventions	1. Cannabis and Psychosis Therapy: mean no. of CAP sessions 8. N=23 2. Psychoeducation: mean no. of sessions 8. N=24.
Outcomes	Behavioural: CASUAS. Mental state: BPRS, SANS, BDI-SF. Social functioning: SOFAS. Global state: KAPQ.
Notes	CAP therapy consisted of a cognitive-behavioural-orientated programme delivered in weekly sessions by trained clinicians over 3 months Psychoeducation was an active control.
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Low risk	Computer generated randomisation.
Allocation concealment?	Unclear risk	Insufficient details.
Blinding? All outcomes	High risk	Single blind.
Incomplete outcome data addressed? All outcomes	High risk	Study attrition not reported.
Free of selective reporting?	Unclear risk	No details.
Free of other bias?	Unclear risk	No details.

Methods	Allocation: randomised computer generated by block using sealed envelopes. Blinding: not stated. Setting: community, Cologne, Bonn, Dusseldorf, and Munich. Inclusion criteria: EIPS criteria. Exclusion criteria: attenuated of brief limited intermittent psychotic symptoms; present or past diagnosis of schizophrenia; alcohol or drug dependence. Follow-up: 12 months therapy.
Participants	Diagnosis: people at risk of developing 1st episode of psychosis. N=128*. Age: mean age ~26 years. Sex: male and female. History: not reported.
Interventions	1. Cognitive Behavioural Therapy: sessions over 12 months, group therapy session n=15, individual therapy session n=25, plus cognitive remediation 12 sessions. N=54. 2. Supportive Therapy: sessions over 12 months, minimal support involving psychoeducation and counselling. N=59
Outcomes	Social functioning: SAS II. Unable to use - Leaving the study early - no usable data.
Notes	*15 participants not accounted for after randomisation. EIPS - early initial prodromal state
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Low risk	Randomised by computer generated numbers.
Allocation concealment?	Unclear risk	Using sealed envelopes.
Blinding? All outcomes	High risk	Not stated.
Incomplete outcome data addressed? All outcomes	High risk	Study attrition only part reported.
Free of selective reporting?	Unclear risk	No details.
Free of other bias?	Low risk	Funded by the German Federal Ministry of Education and Research

Methods	Allocation: randomised (stratified according to affective and non-effective psychosis, by independent statistician). Blinding: single blind. Setting: participant’s home, neutral location or EPPIC centre (Psychosis Prevention and Intervention Centre) Melbourne, Australia. Follow-up: one year. Inclusion criteria: Exclusion criteria: IQ < 70, psychosis due to a medical condition; exhibiting violent behaviour. Evaluation: research assistants blind to treatment.
Participants	Diagnosis: psychosis. N=62. Age: mean 22 years. Sex: 45 M, 17 F. History: mean onset age of psychosis 22 years.
Interventions	1. Cognitive Behavioural Therapy (ACE): dose 20 sessions* for 45 minutes, plus neuroleptics. N=31. 2. Befriending, plus neuroleptics. N=31.
Outcomes	Leaving the study early. Suicide. Hospital admission (not hospitalised). Social functioning: Social and Occupational Functioning Assessment Scale (SOFRAS).
Notes	*Participants could receive a maximum of 20 sessions over 14 weeks for approximately 45 minutes. ACE - Active Cognitive Therapy for Early Psychosis.
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Unclear risk	Randomised, no further details.
Allocation concealment?	Low risk	Randomised by independent statistician.
Blinding? All outcomes	High risk	Single, untested.
Incomplete outcome data addressed? All outcomes	Low risk	Study attrition reported.
Free of selective reporting?	Unclear risk	No details.
Free of other bias?	Low risk	Funded by NH&MRC grant.

Methods	Allocation: randomised by block and placed into sealed envelopes, and drawn by individual with no connection to the study. Blinding: single blind. Setting: outpatients, London, UK. Follow-up: 9 months. Inclusion criteria: people with a 1st episode psychosis within the last 6 months. Exclusion criteria: organic disorders, or learning difficulties. Evaluation: assessment made by research assistants blind to allocation
Participants	Diagnosis: first-episode psychosis (ICD-9). N=106 patients and their carers*. Age:16 + years. Sex: 68 M, 38 F. History: 1st episode psychosis, recruited from psychiatric services in North London
Interventions	1. Brief intervention and treatment as usual. N=57. 2. Treatment as usual with usual support from psychiatric services. N=49. Brief intervention was provided over seven sessions, lasting about one hour and included information gathering from the relative, an educational component on psychotic illness, symptoms and early warning signs, treatment, and help seeking; coping strategies, problem solving and communication with the patient
Outcomes	Leaving the study early. Hospital admission. Unable to use - Satisfaction: Verona Service Satisfaction Questionnaire (no usable data). Perceived severity of illness.
Notes	*Carers/relatives were blind to treatment allocation. ITT analysis used.
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Unclear risk	Randomised, no further details.
Allocation concealment?	Low risk	Using sealed envelopes, drawn by individuals with no connection to the study
Blinding? All outcomes	Low risk	Single, untested.
Incomplete outcome data addressed? All outcomes	Low risk	Study attrition reported.
Free of selective reporting?	Unclear risk	No details.
Free of other bias?	Low risk	Funded by NHS Executive, London Research and Development Programme

Methods	Allocation: cluster randomised with 46 GP practices. Blinding: open study. Setting: community health centre, London, UK. Follow-up: over 12 months.
Participants	Diagnosis: psychosis (CAARMS*) N=113. Age: mean age 23 years. Sex: 81 M, 32 F. Excluded: participants with a history of contact with mental health services for psychosis for more than 6 months or antipsychotic treatment for more than a month. History: first-episode psychosis.
Interventions	1. GP education in early detection with access to LEO-CAT. N=23 practices, N=50 participants. 2. Usual care, with access to LEO-CAT. N=23 practices, N=63 participants
Outcomes	Hospitalisation. Referred to Mental Health Services by Accident and Emergency or Emergency Health services
Notes	*Comprehensive Assessment of At-Risk Mental States (CAARMS) (Yung 2005). LEO-CAT- Lambeth Early Onset Crisis Assessment Team. GP practices randomised into the intervention group received both the GP education training and direct access to the LEO CAT team for referrals.
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Unclear risk	Randomised, no further details.
Allocation concealment?	Unclear risk	No details.
Blinding? All outcomes	High risk	Open study.
Incomplete outcome data addressed? All outcomes	High risk	Study attrition not reported.
Free of selective reporting?	Unclear risk	No details.
Free of other bias?	Unclear risk	No details.

Methods	Allocation: randomised (no further details). Blinding: single, no further details. Setting: Early Psychosis Prevention and Intervention Centre (EPPIC). Follow-up: 10 weeks and 6 months. Inclusion criteria: scoring from 4 to 7 on the expanded version of the BPRS suicidality subscore. Exclusion criteria: attended the EPPIC centre for more than one year. Evaluation: ‘conducted blind to therapy’.
Participants	Diagnosis: first-episode psychosis and acutely suicidal. N=56. Age: range 15-29 years. Sex: not reported. History: not reported.
Interventions	1. LifeSPAN therapy: dose 8 to 10 sessions + standard clinical care. N=31 2. Standard care. N=25. LifeSPAN is a brief individual cognitively orientated therapy specifically designed for acutely suicidal youths with severe mental illness
Outcomes	Leaving the study early. Death from suicide. Unable to use - Mental state: BPRS, SANS: (no usable data). Global state: GAF: (no usable data). Quality of life: (no usable data). Beck Hopelessness Scale: (no usable data). Self Esteem Scale: (no usable data). Self Report Problem Solving Rating Scale: (no usable data). Suicide Ideation Questionnaire: (no usable data). Suicide Intent Scale: (no usable data). Reasons for Living Inventory: (no usable data).
Notes
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Unclear risk	Randomised, no further details.
Allocation concealment?	Unclear risk	No details.
Blinding? All outcomes	High risk	Single blind.
Incomplete outcome data addressed? All outcomes	Low risk	Study attrition reported.
Free of selective reporting?	Unclear risk	No details.
Free of other bias?	Low risk	Funded by the Government of Australia.

Methods	Allocation: ‘randomly assigned’. Blinding: unclear if raters blind to treatment condition. Setting: In-patient unit for adolescents, Amsterdam, Netherlands. Follow-up: 12 months (following on from initial 3 month inpatient admission), then 5 years. Inclusion criteria: first episode of schizophrenia, age 15-26, living or in close contact with parents or other relatives, Dutch speakers, no primary drug problem. Evaluation: ‘independent raters’.
Participants	Diagnosis: schizophrenia. N=76. Age: mean 20.6. Sex: M 53 F 23. History: mean DUP 5.4 months.
Interventions	1. Behavioural family therapy + individual-orientated therapy. N=37 2. Individual-orientated therapy alone. N=39.
Outcomes	Relapse. Unable to use - Mental state: BPRS: (data not reported). Compliance: (data not reported). Lost to follow up: (exact figures unclear, though no evidence of substantial loss)
Notes	First Episode Trial - care from a specialised team plus phase specific intervention versus care from specialised team Unclear when randomisation took place, possible the initial sample size was 97, in which case not intention to treat Five year data reported for whole sample, not by group allocation
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Unclear risk	Randomised, no further details.
Allocation concealment?	Unclear risk	No details.
Blinding? All outcomes	High risk	No details.
Incomplete outcome data addressed? All outcomes	High risk	Study attrition not reported.
Free of selective reporting?	Unclear risk	No details.
Free of other bias?	Unclear risk	Part funded by ‘Praeventiefonds’.

Methods	Allocation: randomised (computer computer-generated ratio of 1:1 in blocks of 6, and stratified for each of 5 centres. In Aarhus, the researchers contacted a secretary by telephone when they had finished the entry assessment of each patient. The secretary then drew 1 lot from among 5 red and 5 white lots out of a black box. When the block of 10 was used, the lots were redrawn. Block sizes were unknown to the investigators). Blinding: raters not blind to treatment allocation, but independent of study group; raters at the 5-year follow-up were blinded to patients’ previous treatment allocation. Setting: Copenhagen, Denmark, multicentre, 5 centres, participants visited in their homes or other places in the community, or at their primary team members office. Exclusion criteria: taking antipsychotics for more than 12 weeks. Duration: 2 years with 5 year follow-up. Evaluations: by independent investigators, not blinded to treatment allocation
Participants	Diagnosis: first-episode schizophrenia spectrum disorder (ICD 10, codes in the F2 category). N=547. Age: range 18-45 years. Sex: M 256, F 291. History: participants included in and outpatients who had not received antipsychotics for more than 12 weeks continuously
Interventions	1. Integrated treatment. N=275. 2. Treatment as usual. N=272. Integrated treatment is an assertive community treatment, enhanced by better specific content via family involvement and social skill training. Treatment as usual consisted of care at a community mental health centre. The treatment was carried out for two years and patients follow up for a further 3 years All participants were offered antipsychotic drugs according to guidelines from the Danish Psychiatric Society, which recommends low dose, atypical antipsychotic strategy for 1st episodes of psychotic illness
Outcomes	Leaving the study early. Global state: GAF. Client Satisfaction: CSQ-8. Suicide attempts. Social outcome: Social Network Schedule, not living independently; no working or in education. Service utilisation: average number of days in hospital. Compliance with treatment. Unable to use - Mental state: SAPS, SANS: (no usable data).
Notes
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Low risk	Randomised, computer generated.
Allocation concealment?	Low risk	Secretary drew lots, which the investigators had no knowledge
Blinding? All outcomes	Low risk	Independent assessors aware of treatment allocation; 5-year follow-up assessors unaware of treatment allocation
Incomplete outcome data addressed? All outcomes	Low risk	Study attrition reported.
Free of selective reporting?	Unclear risk	No details.
Free of other bias?	Low risk	Grants from the Danish Ministry of Health, Danish Ministry of Social Affairs, University of Copenhagen, Copenhagen Hospital Cooperation, Danish Medical Research Council, Wørners Foundation, and the Stanley Wada Research Foundation

Methods	Allocation: simple randomisation by study co-ordinator. Blinding: independent rater, not blind. Setting: PACE clinic (Personal Assessment and Crisis Evaluation), part of EPPIC program, Melbourne, Australia. Inclusion criteria: age 14-30, living in Melbourne, met one of 3 criteria for an Ultra High Risk mental state. Follow-up: 0, 6, 12 months.
Participants	Diagnosis: ‘ultra high risk’ of developing psychosis.* N=59. Age: mean 20. Sex: M 34, F 25. History: not reported.
Interventions	1. Specific Preventive Intervention: dose risperidone 1-2 mg/day + cognitive behavioural therapy + needs-based case management + supportive psychotherapy. N=31 2. Needs based intervention alone. N=28.
Outcomes	Progressing to psychosis. Mental state: BPRS, HRSA, HRSD, SANS, YMS. Quality of Life: QLS. Overall functioning: GAF.
Notes	Prodromal Trial - Care from a specialised team plus a phase specific intervention versus care from a specialised team * Defined as either: family history of psychotic disorder & non specific symps & decrease in functioning on GAF of 30 points or more in last 12 ms, or attenuated psychotic symptoms sustained for at least 1 week, or brief episodes of psychotic symptoms not sustained beyond a week
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Unclear risk	Randomised, no further details.
Allocation concealment?	Unclear risk	No details.
Blinding? All outcomes	High risk	Not blinded.
Incomplete outcome data addressed? All outcomes	High risk	Study attrition not reported.
Free of selective reporting?	Unclear risk	No details.
Free of other bias?	Unclear risk	Funding by the Commonwealth Government of Australia Research and Development Grants Advisory Committee, and Janssen-Cilag Pharmaceuticals; Australian Rotary Health Research Fund grant

Methods	Allocation: randomised. Blinding: assessors blind to treatment group. Setting: no details. Inclusion criteria: 16 years or older, DSM -IV schizophrenia, schizoaffective disorder or schizophreniform disorder and in treatment of a first episode of less than 12 months. Follow-up: 6 months. Evaluation: 3 and 6 months.
Participants	Diagnosis: schizophrenia first episode (DSM IV). N=24. Age: no details. Sex: men and women. History: no details.
Interventions	1. Adherance Coping Education* (ACE) with usual care. N=13. 2. Supportive therapy* with usual care. N=11.
Outcomes	Leaving the study early. Mental state: PANSS, CDPS, CDRS. Quality of life: Heinrichs-Carpenter.
Notes	ACE 14 sessions lasting between 30-45 minutes over six months. It is a manual-based psychotherapy consisting of 4 phases: (1) establishing therapeutic alliance; (2) promoting treatment adherence; (3) developing a plan for maintenance treatment and (4) rehabilitation. Supportive therapy had two phases (1) establishment of the therapeutic relationship and (2) provide emotional support and discussion of non-illness issues or topics. A modified ITT analysis was used based on 19 participants.
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Unclear risk	Randomised, no further details.
Allocation concealment?	Unclear risk	No details.
Blinding? All outcomes	High risk	Not blinded.
Incomplete outcome data addressed? All outcomes	Low risk	Study attrition reported.
Free of selective reporting?	Unclear risk	No details.
Free of other bias?	Unclear risk	No details.

Study	Reason for exclusion
Addington 1999	Allocation: not randomised, before and after design.
Agius 2007	Allocation: not randomised, no control group.
Alanen 1994	Allocation: not randomised, before and after design.
Albiston 1998	Allocation: not randomised, before and after design with historical control
Anonymous 1987	Allocation: randomised. Participants: people with first-episode psychosis. Intervention: medication only (pimozide versus flupenthixol) at standard doses without specific early intervention protocol
Bao 2005	Allocation: randomised. Participants: people with psychosis. Interventions: no early intervention programme.
Birchwood 1989	Allocation: not randomised, service description. Participants: not first-episode patients (study of early detection of signs of relapse)
Clare 1994	Allocation: not randomised, service description. Participants: not first-episode patients (study of early signs of depression in long term patients)
COPE-Melbourne	Allocation: non-randomised quasi-experimental design, controls selected from a similar location to the experimental site
Craig 2004b	Allocation: randomised. Participants: people with 1st & 2nd episode psychosis.
Crow 1986	Allocation: randomised. Participants: people with first-episode psychosis. Intervention: neuroleptic medication at standard doses versus no medication, no specific early intervention protocol
Culberg 1998	Allocation: not randomised - before and after design with historical controls
Davidson 2004	Allocation: randomised. Participants: people with early psychosis. Interventions: atypical versus conventional antipsychotic.
DeHaan 1997	Allocation: not randomised - before and after design.
Drury 2000	Allocation: randomised. Participants: not people with first-episode psychosis.
Emsley 1999	Allocation: randomised. Participants: people with first-episode psychosis. Intervention: medication only (risperidone versus haloperidol), no specific early intervention protocol
Emsley 2004	Allocation: randomised. Participants: people with recent onset schizophrenia. Interventions: medication only (risperidone versus haloperidol)
Falloon 1992	Allocation: not randomised, no controls.
Fisher 2001	Allocation: not randomised - service description, outcome assessed by qualitative survey
Fitzgerald 1998	Allocation: not randomised - before and after design.
Fresan 2001	Allocation: not randomised - before and after study.
Gaebel 2004	Allocation: randomised. Participants: people with first-episode schizophrenia. Interventions: medication only (risperidone versus low-dose haloperidol)
Grawe 1998	Allocation: randomised. Participants: people with first-episode psychosis. Interventions: no early intervention programme.
Hartmann 1974	Allocation: not randomised - retrospective study.
Heydebrand 2004	Allocation: randomised. Participants: people with first-episode schizophrenia. Interventions: haloperidol versus risperidone.
Jenner 2001	Allocation: not randomised - no control group. Participants: not people with first-episode psychosis (adolescents, but on average in treatment for about 3 years)
Jenner 2004	Allocation: randomised. Participants: people with chronic psychosis.
Jolley 2003	Allocation: randomised. Participants: people with 1st & 2nd episode psychosis.
Jones 2005	Allocation: randomised. Participants: people with psychosis - not first episode.
Kadota 1992	Allocation: not randomised - uncontrolled follow-up study of response to neuroleptic treatment
Kauranen 2000	Allocation: not randomised - uncontrolled follow-up study.
Kavanagh 2004	Allocation: randomised. Participants: people with 1st and 2nd episode psychosis.
Keefe 2000	Allocation: randomised. Participants: people with first-episode psychosis. Intervention: medication only (olanzapine versus haloperidol) at standard doses without specific early intervention protocol
Keshavan 1998	Allocation: not randomised - before and after study with historical control group
Kopala 2003	Allocation: randomised. Participants: people with recent onset schizophrenia. Interventions: risperidone versus haloperidol.
Kuipers 2004	Allocation: randomised. Participants: people with schizophrenia, 1st and 2nd or more episodes
Li 2004	Allocation: randomised. Participants: early schizophrenia, probably not 1st episode. Inteventions: no early intervention program.
Lieberman 2005b	Allocation: randomised. Participants: people with first-episode psychosis and healthy volunteers. Interventions: medication only (haloperidol versus olanzapine)
Malla 2001	Allocation: not randomised - before and after study, no control group
McCay 2007	Allocation: quasi-randomised.
McGorry 1996	Allocation: not randomised - before and after study with historical controls
Mosher 1975	Allocation: not randomised - allocation on ‘a consecutively admitted, space available basis’ to ‘Soteria’ - a small home like facility in the community which acted as an alternative to admission for patients in their first-episode of schizophrenia
Mottaghipour 2000	Allocation: not randomised, no control group (participants compared with a group of families of long term patients)
Newton 2005	Allocation: none randomised study (before and after design).
Nuechterlein 2005	Allocation: randomised. Participants: people with first-episode psychosis. Interventions: no early intervention programme.
Parlato 1999	Allocation: non-randomised - a description of a service.
Perez 2003	Allocation: randomised. Participants: people with first-episode psychosis. Interventions: medication only (olanzapine versus risperidone versus haloperidol)
Power 2004	Allocation: randomised. Participants: people with 1st and 2nd episode psychosis.
Purdon 2000	Allocation: randomised. Participants: people with first-episode psychosis. Intervention: medication only (olanzapine versus risperidone versus haloperidol) without specific early intervention protocol
Rund 1994	Allocation: not randomised, before and after study with historical control
Sanger 1999	Allocation: randomised. Participants: people with first-episode psychosis. Intervention: medication only (olanzapine versus haloperidol) without specific early intervention protocol
Schooler 2003	Allocation: randomised. Participants: people with recent onset schizophrenia. Interventions: medication only (risperidone versus haloperidol)
SOCRATES-UK	Allocation: randomised. Participants: not first-episode patients, participants could be in either first or second admission, as long as second admission within 2 years of first admission (estimated that 61/309 participants were not first episode)
Szymanski 1994	Allocation: not randomised - before and after study without control group
Thomas 1979	Allocation: not randomised - no control group. Participants: not first-episode patients, although all were adolescents, some were experiencing an exacerbation of chronic schizophrenia
TIPS 2006	Allocation: not randomised.
Turetz 1997	Allocation: not randomised - no control group. Participants: probably not first-episode patients, although all participants were children, they were selected on the basis of treatment resistance and so probably not in the first episode
Ueland 2004	Allocation: randomised. Participants: people with psychosis, not first episode.
Walczewski 1998	Allocation: not randomised - a quasi-experimental design, patients receiving a psychosocial treatment program were compared with a group receiving an individual treatment programme
Wang 2000	Allocation: randomised. Participants: people with first-episode psychosis. Intervention: medication only (risperidone versus clozapine) without specific early intervention protocol
Welch 2000	Allocation: not randomised - service description.
Whitehorn 1998	Allocation: not randomised - before and after study without control
Whitwell 2000	Allocation: not randomised - service description.
Wieneke 2000	Allocation: not randomised - service description.
Wunderink 2003	Allocation: randomised. Participants: people with early onset psychosis. Intervention: drug trial - not early intervention.
Wykes 2007	Allocation: randomised. Participants: people with psychosis, but not first episode, (only early onset)
Yap 2001	Allocation: not randomised - before and after study without control group
Zhang-Wong 1999	Allocation: not randomised - prospective uncontrolled study to determine optimal dose of haloperidol

Methods	Randomised.
Participants	People with psychosis.
Interventions	Early intervention.
Outcomes	No data.
Notes

Methods	Unclear.
Participants	People with psychosis.
Interventions	Unclear.
Outcomes	Unclear.
Notes

Methods	Controlled trial.
Participants	People with early psychosis.
Interventions	Cognitive and family therapy relapse prevention.
Outcomes	No details.
Notes

Methods	Cluster randomised trial.
Participants	People with early psychosis.
Interventions	Specialist team versus augmented community mental health teams
Outcomes	No details.
Notes

PERMALINK

Early intervention for psychosis

Max Marshall

John Rathbone

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors’ conclusions

BACKGROUND

Description of the condition

Description of the intervention

How the intervention might work

Why it is important to do this review

OBJECTIVES

METHODS

Criteria for considering studies for this review

Types of studies

1. Trials to prevent the development of psychosis

2. Trials to improve outcome in first-episode psychosis

Types of participants

Types of interventions

1. Trials to prevent the development of psychosis in prodromal patients

1.1 Phase-specific treatment

1.2 Care from a specialised team

1.3 Control conditions

2. Trials to improve outcome in first-episode psychosis

2.1 Early detection

2.2 Phase-specific treatment

2.3 Care from a specialised team

2.4 Control conditions

3. Excluded interventions

Types of outcome measures

Primary outcomes

1. General

1.1 Converting to psychosis during follow-up period

1. General

1.1 Relapse - as defined by each study or re-admission during follow-up

Secondary outcomes

1. General

2. Mental state

3. Behaviour

4. Adverse effects

5. Economic

6. Quality of life

1. General

2. Mental state

3. Behaviour

4. Adverse effects

5. Economic

Search methods for identification of studies

Electronic searches

1. Cochrane Schizophrenia Group Trials Register (March 2009)

2. Previous searches for earlier versions of this review

Searching other resources

1. Reference lists

2. Personal contact

Data collection and analysis

Selection of studies

Data extraction and management

1. Extraction

2. Management

2.1 Forms

2.2 Direction of graphs

2.3 Scale-derived data

2.4 Skewed data

2.5 Final endpoint value versus change data

2.6 Common measure

2.7 Conversion of continuous to binary

2.8 Summary of findings table

Assessment of risk of bias in included studies

Measures of treatment effect

1. Binary data

2. Continuous data

Unit of analysis issues

1. Cluster trials

2. Cross-over design

3.2 Employing the I² statistic

Trial name or title	A randomised controlled trial of individual therapy for first-episode psychosis
Methods	Randomised.
Participants	People with schizophrenia.
Interventions	CBT versus usual care.
Outcomes	Social Functioning Scale (SFS) Positive and Negative Syndrome Scale (PANSS) Psychotic Symptom Rating Scales (PSYRATS) Calgary Depression Scale for Schizophrenia (CDSS) The Time-Line Follow Back (TLFB) Alcohol and Drug Use Scale (AUS; DUS) Medication Event Monitoring System (MEMS) Rosenberg Self-Esteem Scale Maastricht Assessment of Coping Skills (MACS)
Starting date	June 2007.
Contact information	Jean_Addington@camh.net
Notes

Trial name or title	Prodromal symptoms and early intervention to prevent a relapse
Methods	No details.
Participants	People with schizophrenia.
Interventions	The Symptom Management Module (SMM) versus treatment as usual
Outcomes	Occurrence of a psychotic relapse: a worsening of at least two points on the CGI as assessed by psychiatrist and verified by researcher by a PANSS-interview within a week.
Starting date	No details.
Contact information	Johan.Arends@GGZDrenthe.nl
Notes

Trial name or title	Early detection and psychological intervention for individuals at high risk of psychosis (EDIE 2)
Methods	Randomised.
Participants	People with high risk of psychosis.
Interventions	CBT plus regular monitoring versus monitoring alone.
Outcomes	Transition to psychosis.
Starting date	No details.
Contact information	No details.
Notes

Trial name or title	Access, detection and psychological treatments.
Methods	No details.
Participants	People at risk of developing psychosis.
Interventions	Psychological intervention in preventing or delaying the onset of a psychotic illness.
Outcomes	Psychosis.
Starting date	August 2004.
Contact information	No details.
Notes

Trial name or title	Maintenance treatment versus stepwise drug discontinuation in first-episode schizophrenia
Methods	Randomised.
Participants	People with schizophrenia.
Interventions	Maintenance treatment.
Outcomes	Relapse.
Starting date	November 2001.
Contact information	No details.
Notes

Trial name or title	Sarcosine preventive therapy for individuals at high risk for schizophrenia
Methods	Randomised.
Participants	People with the prodromal stage of schizophrenia.
Interventions	Sarcosine versus placebo.
Outcomes	Delay or prevention of illness.
Starting date	June 2006.
Contact information	No details.
Notes

Trial name or title	Birmingham early detection in untreated psychosis trial (REDIRECT)
Methods	Cluster randomised.
Participants	People with first-episode psychosis, aged 14-30 years.
Interventions	Early detection training versus detection as usual.
Outcomes	The primary outcome is the number of general practitioner referrals of young people with first-episode psychosis to early intervention services. Secondary outcomes are duration of untreated psychosis, time to recovery, use of the Mental Health Act, and general practitioner consultation rate
Starting date	2004.
Contact information	No details.
Notes

Trial name or title	Early detection and intervention prevention of psychosis.
Methods	Randomised, single blind (outcomes assessor).
Participants	People with a high risk of psychosis.
Interventions	Psychoeducational multifamily group treatment plus antipsychotics versus case management
Outcomes	Conversion to psychosis.
Starting date	No details.
Contact information	mcfarw@mmc.org
Notes

Trial name or title	Specialized treatment early in psychosis (STEP): a pragmatic randomized controlled trial in the US public sector
Methods	Randomised.
Participants	People with early psychosis.
Interventions	Specialized treatment for early psychosis.
Outcomes	No details.
Starting date	No details.
Contact information	No details.
Notes

Trial name or title	The depth project: a multi site RCT for youths at risk for psychosis
Methods	Randomised.
Participants	People at risk of psychosis.
Interventions	Cognitive behavioural therapy and person centred therapy in ameliorating ‘at risk mental states’ for psychosis
Outcomes	No details.
Starting date	No details.
Contact information	No details.
Notes