Intervention Review

You have free access to this content

Pimozide for schizophrenia or related psychoses

  1. Meghana Mothi1,*,
  2. Stephanie Sampson2

Editorial Group: Cochrane Schizophrenia Group

Published Online: 5 NOV 2013

Assessed as up-to-date: 17 MAY 2007

DOI: 10.1002/14651858.CD001949.pub3


How to Cite

Mothi M, Sampson S. Pimozide for schizophrenia or related psychoses. Cochrane Database of Systematic Reviews 2013, Issue 11. Art. No.: CD001949. DOI: 10.1002/14651858.CD001949.pub3.

Author Information

  1. 1

    Newsam Centre, Seacroft Hospital, General Adult Psychiatry, Leeds, UK

  2. 2

    The University of Nottingham, Cochrane Schizophrenia Group, Nottingham, UK

*Meghana Mothi, General Adult Psychiatry, Newsam Centre, Seacroft Hospital, Leeds and York NHS Foundation Trust, York Road, Leeds, LS14 6WB, UK. m.mothi@nhs.net.

Publication History

  1. Publication Status: New search for studies and content updated (conclusions changed)
  2. Published Online: 5 NOV 2013

SEARCH

 

Summary of findings    [Explanations]

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms

 
Summary of findings for the main comparison. PIMOZIDE versus PLACEBO for schizophrenia or related psychoses

PIMOZIDE versus PLACEBO for schizophrenia or related psychoses

Patient or population: patients with schizophrenia or related psychoses
Settings: Inpatient and outpatient (UK, USA)
Intervention: PIMOZIDE versus PLACEBO

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments

Assumed riskCorresponding risk

PlaceboPIMOZIDE

Global state: relapse medium term (3-12 months)
Clinical diagnoses
Follow-up: 1 year
375 per 1000 182 per 1000
(11 to 667)
RR 0.22
(0.03 to 1.78)
20
(1 study)
⊕⊝⊝⊝
very low2,3

Mental state: no improvement medium term (3-12 months) not reportedSee comment.See comment.Not estimable-See comment.No study reported this outcome.

Mental state: specific symptoms first-rank symptoms medium term (3-12 months) not reportedSee comment.See comment.Not estimable-See comment.No study reported this outcome.

Adverse effects: extrapyramidal adverse effects: Parkinsonism (rigidity) short term (<3 months)
Clinical diagnoses
Follow-up: 12 weeks
0 per 100040 per 1000
(0 to 0)
RR 5.5
(0.3 to 101.28)
19
(1 study)
⊕⊝⊝⊝
very low3,5,6

Adverse effects: extrapyramidal adverse effects:Parkinsonism (rigidity) medium term (3-12 months)
Clinical diagnoses
Follow-up: 6 months
100 per 10001133 per 1000
(14 to 1000)
RR 1.33
(0.14 to 12.82)
25
(1 study)
⊕⊝⊝⊝
very low3,6,7

Adverse effects: extrapyramidal adverse effects:Parkinsonism (tremor) medium term (3-12 months)
Clinical diagnoses
Follow-up: 6 months
200 per 10001200 per 1000
(40 to 990)
RR 1
(0.2 to 4.95)
25
(1 study)
⊕⊝⊝⊝
very low3,6,7

Quality of life: no significant change in quality of life/satisfaction medium term (3-12 months) not reportedSee comment.See comment.Not estimable-See comment.No study reported this outcome.

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio.

GRADE Working Group grades of evidence:
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1Mean baseline risk presented for single study.
2Indirectness: 'serious'; comparison of pimozide vs placebo was made during a six-month follow-up period, in which all injections (both active and placebo) were withdrawn from a selection of participants who had remained well throughout the previous 12-month period of the trial.
3Imprecision: 'very serious'; small sample size.
4Mean baseline risk: only one trial reported with 0 events in the control group and 2 events in the intervention group.
5Risk of bias: 'serious'; not all participants from the study accounted for.
6Imprecision: 'very serious'; confidence intervals for best estimate of effect include both 'no effect' and appreciable benefit/harm.
7Risk of bias: 'serious'; raters not independent of treatment in this small, single study.

 Summary of findings 2 PIMOZIDE versus ANY ANTIPSYCHOTIC for schizophrenia or related psychoses

 Summary of findings 3 PIMOZIDE + ANY ANTIPSYCHOTIC versus ANY ANTIPSYCHOTIC for schizophrenia or related psychoses

 Summary of findings 4 PIMOZIDE + ANY ANTIPSYCHOTIC versus ANTIPSYCHOTIC + PLACEBO for schizophrenia or related psychoses

 Summary of findings 5 PIMOZIDE + ANY ANTIPSYCHOTIC versus ANTIPSYCHOTIC + ANTIPSYCHOTIC for schizophrenia or related psychoses

 

Background

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms
 

Description of the condition

Schizophrenia is a severe and enduring mental illness with a lifetime risk of about 1% (Davies 2007). Symptoms of schizophrenia are commonly clustered into positive, negative and cognitive symptoms. Positive symptoms comprise hallucinations, delusions and thought disorder. Negative symptoms sometimes are more difficult to elicit; however, it is generally accepted that apathy, avolition, anhedonia, alogia and impairment of attention are negative symptoms. Cognitive decline can happen over many years and can be difficult to detect (Radhakrishnan 2012). Often the illness can present with a mixture of the above cluster of symptoms. Growing concern surrounds the fact that people who suffer from schizophrenia are more likely to have higher morbidity from physical illness compared with the general population (De Hert 2011).

Delusions are commonly defined as false fixed beliefs that are out of keeping with the person's cultural background. Delusional beliefs can occur as part of the cluster of positive symptoms of schizophrenia and can take the form of a simple or elaborate or a well-systematised set of beliefs. Delusional disorders are defined as 'ICD-10 definition' (ICD 10 1994). Antipsychotics have been the mainstay of treatment for schizophrenia, and anecdotally pimozide has been used most often to treat delusional disorders. Pimozide was one of the earlier antipsychotic medications that was thought to have an association with sudden cardiac death. In the UK, the British National Formulary (BNF) recommends monitoring of ECT before treatment initiation, and if evidence of QT interval prolongation is found, withdrawing the treatment or reducing the dose and reviewing the medication. Of course more recently, other antipsychotics have come under the scanner as well, and QT prolongation is being recognised increasingly as a possible side effect of these drugs.

 

Description of the intervention

Pimozide was first formulated in the late 1960s by Janssen Pharmaceutica (Janssen 1998; Udabe 1998; Yaryura-Tobias 1998) and was marketed for the care of people with schizophrenia or related psychoses such as delusional disorder. Pimozide is often advocated as the preferred drug treatment for delusional disorder (Kaplan 1994; Opler 1995), and it has been claimed that it is particularly effective for one subtype, namely, monosymptomatic hypochondriacal psychosis (Munro 1980; Ungvari 1986). Pimozide has also been claimed to have particular efficacy for negative symptoms such as apathy, poor volition and slow and 'impoverished' thinking (Feinberg 1988; Opler 1991).

Pimozide is generally well tolerated apart from extrapyramidal side effects (tremor, muscle stiffness and slow body movements). However, it has been associated with cardiotoxicity and sudden unexplained death. Electrocardiogram monitoring is required before and during its use (BNF 2012), and pimozide is now contraindicated for people with known heart problems (prolonged QT interval) or a history of arrhythmia (BNF 2012).

 

How the intervention might work

Pimozide is a diphenylbutylpiperidine antipsychotic drug (1-[1-[4,4-bis(p-fluorophenyl)butyl]-4-piperidyl]-2-benzimidazolinone; Figure 1). It is an oral antipsychotic drug that differs from other antipsychotic drugs in several ways. It has a long half-life (55 to 150 hours), which means that it can be given as infrequently as once daily in acute psychotic illness and three times weekly in the maintenance phase (Wilson 1982*; King 1995). This may be useful for people whose compliance is poor but who refuse depot medication (McCreadie 1987).

 FigureFigure 1. Pimozide structure.

Pimozide is derived from the butyrophenone class, of which haloperidol is the best known example. It differs from haloperidol in that it has a more specific dopamine blocking action (D2 receptor antagonist with very high affinity for 5-HT7 receptor) and is therefore less sedative (Silverstone 1995). It is distinguished from other antipsychotic drugs by its potent calcium blocking activity. The usual dose ranges from 2 to 20 mg daily, with a maximum licensed daily dose for adults of 20 mg (BNF 2012).

 

Why it is important to do this review

Pimozide is a drug that was developed in the 1960s for the treatment of schizophrenia or related psychoses, and it remains available in the formulary (BNF) for treating patients with schizophrenia and monosymptomatic hypochondriacal psychosis. It is important that we gather up-to-date evidence for this drug to review its effectiveness and adverse effect profile.

 

Objectives

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms

To review the effects of pimozide for people with schizophrenia or related psychoses in comparison with placebo, no treatment or other antipsychotic medication.

A secondary objective was to examine the effects of pimozide for people with delusional disorder.

 

Methods

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms
 

Criteria for considering studies for this review

 

Types of studies

All relevant randomised controlled trials (RCTs). We excluded quasi-randomised studies, such as those allocating by alternate days of the week. If a trial was described as double blind and implied randomisation, we included such trials in a sensitivity analysis (see Sensitivity analysis). Randomised cross-over trials were eligible for inclusion only up to the point of the first cross-over, because of potential carry-over effects of treatments (for previous methods, see Appendix 1).

 

Types of participants

We included people with a diagnosis of schizophrenia, delusional disorder (including monosymptomatic hypochondriacal psychosis), however defined. We also included people with 'serious/chronic mental illness' or 'psychotic illness' and 'dual diagnosis' patients with serious mental illness. If possible, we excluded people with schizoaffective disorder, dementing illnesses, depression and primary problems associated with substance misuse.

 

Types of interventions

 

1. Pimozide (any dose and pattern of administration) versus:

 
1.1 Placebo

 
1.2 Typical antipsychotic

Including chlorpromazine, chlorprothixene, clopenthixol, cyamemazine, droperidol, flupentixol/flupenthixol, fluphenazine, haloperidol, levomepromazine, loxapine, mesoridazine, molindone, periciazine, perphenazine, pimozide, prochlorperazine, promazine, promethazine, thioridazine, thiothixene, trifluoperazine, triflupromazine, zuclopenthixol.

 
1.3 Atypical antipsychotic

Including amisulpride, aripiprazole, asenapine, clozapine, clothiapine, clotiapin, iloperidone, lurasidone, mosapramine, olanzapine, paliperidone, perospirone, quetiapine, remoxipride, risperidone, sertindole, sulpiride, ziprasidone, zotepine.

 

2. Pimozide + any (atypical/typical) antipsychotic versus:

 
2.1. Placebo

 
2.2 Any (atypical/typical) antipsychotics

 

Types of outcome measures

As schizophrenia is often a life-long illness and pimozide is used as an ongoing treatment, we grouped outcomes according to time periods: short term (<3 months), medium term (3 to 12 months) and long term (>1 year).

 

Primary outcomes

 
1. Global state

1.1 Any clinically significant response in global state, as defined by each study.
1.2 Relapse, as defined by each study.
1.3 No improvement, as defined by each study.
1.4 Average score/change in global state.

 
2. Mental state

2.1 No clinically significant response, as defined by each study.
2.2 No improvement, as defined by each study.
2.3 Average score/change in mental state.
2.4 Average score/change in negative symptoms.
2.5 Any clinically significant response in mental state, as defined by each study.
2.6 Any clinically significant response in negative symptoms, as defined by each study.
2.7 Use of additional medication (other than anticholinergics) for psychiatric symptoms.

 
3. Adverse effects: extrapyramidal side effects

3.1 Any clinically significant extrapyramidal adverse effects.
3.2 Any clinically significant extrapyramidal side effects, as defined by each study.
3.3 Average score/change in extrapyramidal side effects.
3.4 Incidence of use of antiparkinson drugs.

 

Secondary outcomes

 
1. Service utilisation outcomes

1.1 Hospital admission.
1.2 Days in hospital.

 
2. Behaviour

2.1 Any clinically significant response in behaviour, as defined by each study.
2.2 Average score/change in behaviour.
2.3 Aggression/violence.

 
3. Adverse effects

3.1 Death.
3.2 Anticholinergic effects.
3.3 Antihistamine effects (such as sedation).
3.4 Cardiovascular effects.
3.5 Central nervous system effects.
3.6 Gastrointestinal effects.
3.7 Dermatological effects.
3.8 Endocrine effects and prolactin-related symptoms.
3.9 Genitourinary effects.
3.10 Any abnormal laboratory tests.
3.11 Any other specific adverse effects.

 
4. Social functioning

4.1 Any clinically significant response in social functioning, as defined by each study.
4.2 No improvement, as defined by each study.
4.3 Average score/change in social functioning.

 
5. Quality of life/satisfaction with care for either recipients of care or carers

5.1 Significant change in quality of life/satisfaction, as defined by each study.
5.2 Average score/change in quality of life/satisfaction.
5.3 Any change in employment status, as defined by each study.

 
6. Cognitive functioning

6.1 Any significant change in cognitive functioning, as defined by each study.
6.2 Degree of change in cognitive functioning, as defined by each study.

 
7. Economic outcomes

7.1 Costs due to treatment, as defined by each study.
7.2 Savings due to treatment, as defined by each study.

 
8. Leaving the study early

8.1 Owing to relapse.
8.2 Owing to adverse effects.
8.3 For any reason.

 
9. Summary of findings table

We used the GRADE approach to interpret findings (Schünemann 2008) and used the GRADE profiler to import data from Review Manager 5 (Revman) to create 'Summary of findings' tables. These tables provide outcome-specific information concerning the overall quality of evidence from each included study in the comparison, the magnitude of effect of the interventions examined, and the sum of available data on all outcomes that we rated as important to patient care and decision making. We selected the following main outcomes for inclusion in the 'Summary of findings' table.

  1. Global state: relapse, as defined in each study (medium term).
  2. Mental state: no improvement, as defined by each study (medium term).
  3. Mental state: specific symptoms, first-rank symptoms (medium term).
  4. Extrapyramidal adverse effects: Parkinsonism rigidity (short term).
  5. Extrapyramidal adverse effects: Parkinsonism rigidity (medium term).
  6. Quality of life: no significant change in quality of life/satisfaction (medium term).

This table provides information concerning the overall quality of the evidence from the trial, the magnitude of effect of the interventions examined, and the sum of available data on all primary outcomes and on selected secondary outcomes. This summary was used to guide our conclusions and recommendations.

As schizophrenia is often a lifelong illness and pimozide is used as an ongoing treatment, we grouped outcomes according to time periods: short term (less than 3 months), medium term (3 to 12 months) and long term (longer than 1 year).

 

Search methods for identification of studies

 

Electronic searches

 

1. Cochrane Schizophrenia Group Trials Register (January 2013)

We searched the register using the phrase:

 [(*pimozide* or * orap* or *antalon* or *opiran* or *pirium* or *R-6238* in REFERENCE title, abstract and index fields) OR (*pimozide* or * orap* or *antalon* or *opiran* or *pirium* or *R-6238*) in STUDY intervention field]

 This register is compiled by systematic searches of major databases, handsearches and searches of conference proceedings (see group module).

 

2. Previous electronic searches

For previous search strategies, see Appendix 2.

 

Searching other resources

 

1. Reference Searching

We inspected references of all included studies for further relevant studies.

 

2. Handsearching

We sought additional relevant trials by handsearching reference lists of included and excluded trials.

 

3. Requests for additional data

We attempted to contact authors of relevant trials to inquire about other sources of relevant information.

 

Data collection and analysis

Methods used in data collection and analysis for this 2013 update are provided below; for previous methods, please see Appendix 3.

 

Selection of studies

For this 2013 update, review authors MM and SS independently inspected citations from the new electronic search and identified relevant abstracts. MM and SS also inspected full articles of the abstracts meeting inclusion criteria. CEA (see Acknowledgements) carried out the reliability check of all citations from the new electronic search. We resolved any disagreement by discussion. Where difficulties or disputes arose, we asked CEA for help and if it was impossible to decide, added these studies to those awaiting assessment and contacted the authors of the papers for clarification.

 

Data extraction and management

 

1. Extraction

For this 2013 update, review authors MM and SS extracted data from all included studies. In addition, to ensure reliability, CEA independently extracted data from a 10% random sample of the total included studies. We discussed any disagreement, documented decisions and, if necessary, contacted authors of studies for clarification. With remaining problems, CEA helped to clarify issues, and we documented those final decisions. We extracted data presented only in graphs and figures whenever possible but included these only if two review authors independently attained the same result. We attempted to contact authors through an open-ended request to obtain missing information or for clarification whenever necessary. Had we encountered cluster trials, we would have extracted data relevant to each component centre of such studies separately.

 

2. Management

 
2.1 Forms

We extracted data onto standard, simple forms.

 
2.2 Scale-derived data

We included continuous data from rating scales only if:
a. the psychometric properties of the measuring instrument had been described in a peer-reviewed journal (Marshall 2000); and
b. the measuring instrument had not been written or modified by one of the trialists for that particular trial.

Ideally, the measuring instrument should be a self-report or should be completed by an independent rater or relative (not the therapist). We realise that often this is not reported clearly; we have noted whether or not this is the case in Description of studies and in the 'Risk of bias' tables.

 
2.3 Endpoint versus change data

Both endpoint and change data offer advantages. Change data can remove a component of between-person variability from the analysis. On the other hand, calculation of change needs two assessments (baseline and endpoint), which can be difficult to perform in unstable and difficult to measure conditions such as schizophrenia. We decided to use primarily endpoint data and to use change data only if the former were not available. We combined endpoint and change data in the analysis, as we used mean differences (MDs) rather than standardised mean differences throughout (Higgins 2011; Chapter 9.4.5.2).

 
2.4 Skewed data

Continuous data on clinical and social outcomes often are not normally distributed. To avoid the pitfall of applying parametric tests to non-parametric data, we aimed to apply the following standards to all data before inclusion: (a) standard deviations and means are reported in the paper or can be obtained from the authors; (b) when a scale starts from the finite number zero, the standard deviation (SD), when multiplied by two, is less than the mean (as otherwise, the mean is unlikely to be an appropriate measure of the centre of the distribution (Altman 1996)); (c) if a scale started from a positive value (such as the Positive and Negative Syndrome Scale (PANSS), which can have values from 30 to 210), we modified the calculation described above to take the scale starting point into account. In these cases, skew is present if 2 SD > (S Smin), where S is the mean score and S min is the minimum score. Endpoint scores on scales often have finite starting and ending points, and these rules can be applied. We entered skewed endpoint data from studies of fewer than 200 participants as other data within the data and analyses section rather than into the statistical analysis. Skewed endpoint data pose less of a problem when one is looking at the mean if the sample size is large; had we come across such larger-scale studies, we would have entered such data into syntheses. When continuous data are presented on a scale that includes a possibility of negative values (such as change data), it is difficult to tell whether or not data are skewed.

 
2.5 Common measure

To facilitate comparison between trials, we intended to convert variables that can be reported in different metrics, such as days in hospital (mean days per year, per week or per month), to a common metric (e.g. mean days per month). However, no such variables were found in the included studies.

 
2.6 Conversion of continuous to binary

Where possible, we made efforts to convert outcome measures to dichotomous data. This can be done by identifying cut-off points on rating scales and dividing participants accordingly into 'clinically improved' and 'not clinically improved'. It is generally assumed that if a 50% reduction is seen in a scale-derived score such as the Brief Psychiatric Rating Scale (BPRS; Overall 1962) or the PANSS (Kay 1986), this could be considered as a clinically significant response (Leucht 2005; Leucht 2005a). If data based on these thresholds were not available, we used the primary cut-off presented by the original authors. Most included studies dichotomised data from rating scales to rate them as 'improved' or 'not improved' (as defined in each study), which was included under the global state outcome of 'no improvement'. Where conversions have not been made, this means that a 50% reduction in the scale-derived data was not found.

 
2.7 Direction of graphs

Where possible, we entered data in such a way that the area to the left of the line of no effect indicated a favourable outcome for pimozide. If we had to enter data in a way that the area to the left of the line indicates an unfavourable outcome, this was noted in the relevant graphs.

 

Assessment of risk of bias in included studies

For this 2013 update, MM and SS worked independently to assess trial quality by using criteria described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). This new set of criteria is based on evidence of associations between overestimate of effect and high risk of bias of the article, such as sequence generation, allocation concealment, blinding, incomplete outcome data and selective reporting.

Where inadequate details of randomisation and other characteristics of trials were provided, we contacted authors of the studies to obtain additional information.

We have noted the level of risk of bias both in the text of the review and in the 'Summary of findings' tables.

 

Measures of treatment effect

 

1. Binary data

For binary outcomes, we calculated a standard estimation of the risk ratio (RR) and its 95% confidence interval (CI). It has been shown that RR is more intuitive (Boissel 1999) than odds ratios and that odds ratios tend to be interpreted as RR by clinicians (Deeks 2000). If heterogeneity was identified (Assessment of heterogeneity), we used a random-effects model. For statistically significant results, we used 'Summary of findings' tables to calculate the number needed to treat for an additional beneficial outcome /for an additional harmful outcome statistic and its 95% CI.

 

2. Continuous data

For continuous outcomes, we estimated the mean difference (MD) between groups. We would prefer to not calculate effect size measures (standardised mean difference (SMD)). However, if scales of very considerable similarity were used, we presumed that a small difference in measurement was present, and we would have calculated effect size and transformed the effect back to the units of one or more of the specific instruments. However, no such data were found.

 

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. Authors often fail to account for intraclass 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).

Had we included any cluster trials, and in cases where clustering was not accounted for in primary studies, we would have presented data in a table, with an (*) symbol to indicate the presence of a probable unit of analysis error. In subsequent versions of this review, if any cluster trials are identified, we will seek to contact first authors of such studies to obtain intraclass correlation coefficients for their clustered data and to adjust for this by using accepted methods (Gulliford 1999).

If clustering had been incorporated into the analysis of primary studies, we would have presented these data as if from a non-cluster randomised study, with adjustment 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 by using the mean number of participants per cluster (m) and the intraclass correlation coefficient (ICC) [Design effect = 1 + (m1) * ICC] (Donner 2002). Again, had we found any cluster trials, or if subsequent versions of this review should identify such trials, when the ICC is not reported it will be assumed to be 0.1 (Ukoumunne 1999).

 

2. Cross-over trials

A major concern of cross-over trials is the carry-over effect. It occurs when an effect (e.g. pharmacological, physiological, psychological) of treatment in the first phase is carried over to the second phase. As a consequence, on entry to the second phase, 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 highly likely in severe mental illness, we would have used only data from the first phase of cross-over studies, had any such studies been identified. However, this review does not include any cross-over design studies.

 

3. Studies with multiple treatment groups

When a study involves more than two treatment arms, if relevant, we presented the additional treatment arms in comparisons. If data were binary, we simply added them and combined them within the two-by-two table. If data were continuous, we combined them by following the formula provided in Section 7.7.3.8 ('Combining groups') of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). When additional treatment arms were not relevant, we did not reproduce these data.

 

Dealing with missing data

 

1. Overall loss of credibility

At some degree of loss of follow-up, data must lose credibility (Xia 2009). We chose that, for any particular outcome, should more than 40% of data be unaccounted for, we did not reproduce these data or use them within analyses. If, however, more than 40% of those in one arm of a study were lost, but the total loss was less than 40%, we marked such data with an (*) to indicate that such a result may well be prone to bias. This was the case in Gross 1974*; Clark 1975*; Abuzzahab 1980*; McCreadie 1982* and Wilson 1982*.

 

2. Binary

When attrition for a binary outcome was between 0% and 60%, and when data were not clearly described, we presented data on a 'once-randomised-always-analyse' basis (an intention-to-treat (ITT) analysis). All ot those leaving the study early were assumed to have the same rates of negative outcome as those who completed, with the exception of the outcomes of death and adverse effects. For these outcomes, the rate of those who stayed in the study in that particular arm of the trial was used for those who did not. We undertook a sensitivity analysis to test how prone the primary outcomes were to change when 'completer' data only were compared with the ITT analysis by using the above assumptions.

 

3. Continuous

 
3.1 Attrition

In cases where attrition for a continuous outcome is between 0% and 50% and completer-only data are reported, we reproduced these data.

 
3.2 Standard deviations

Had we been confronted with missing SDs, we would have tried first to obtain the missing values from the authors. If the missing values were not available, where measures of variance for continuous data were missing but an exact standard error (SE) and CIs were available for group means, and either a P value or a T value was available for differences in means, we were able to calculate them according to the rules described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011): When only the SE is reported, SDs are calculated by the formula SD = SE * square root (n). Chapters 7.7.3 and 16.1.3 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) present detailed formulae for estimating SDs from P values, T or F values, confidence intervals, ranges or other statistics. If these formulae do not apply, we would have calculated the SDs according to a validated imputation method based on the SDs of the other included studies (Furukawa 2006). Although some of these imputation strategies can introduce error, the alternative would be to exclude a given study’s outcome and thus to lose information. However, we did not have to impute any of these values.

 
3.3 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 applied to deal with missing data, LOCF introduces uncertainty about the reliability of the results (Leucht 2007). Therefore, in cases where LOCF data have been used in the trial, if less than 50% of the data have 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 people or situations that we had not predicted would arise. When such situations or participant groups arose, we 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 that we had not predicted would arise. When such methodological outliers arose, we 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 statistical method alongside the Chi2 P value. The I2 statistic 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 the magnitude and direction of effects and the strength of evidence for heterogeneity (e.g. P value from Chi2 test, confidence interval for I2). An I2 estimate greater than or equal to around 50% accompanied by a statistically significant Chi2 statistic was interpreted as evidence of substantial levels of heterogeneity (Higgins 2011). 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

 

1. Protocol versus full study

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 Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We attempted to locate protocols of included randomised trials. If the protocol was available, outcomes in the protocol and in the published report were compared. If the protocol was not available, outcomes listed in the Methods section of the trial report were compared with actually reported results.

 

2. Funnel plot

Reporting biases arise when the dissemination of research findings is influenced by the nature and direction of results (Egger 1997). These are described in Section 10 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We are aware that funnel plots may be useful for investigating reporting biases but are of limited power for detecting small-study effects. We did not use funnel plots for outcomes when 10 or fewer studies were included, or when all studies were of similar size. In other cases, for which funnel plots were possible, we sought statistical advice in their interpretation (see Figure 2).

 FigureFigure 2. Funnel plot of comparison: 3 PIMOZIDE versus ANY ANTIPSYCHOTIC, outcome: 3.3 Global state: 2. No improvement.

 

Data synthesis

We understand that there is no closed argument regarding preference for the 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 often seems to be true to us, and the random-effects model takes into account differences between studies even if no statistically significant heterogeneity is noted. However, the random-effects model confers a disadvantage: It puts added weight onto small studies, which often are the most biased ones. Depending on the direction of effect, these studies can inflate or deflate the effect size. We chose the fixed-effect model for all analyses.

 

Subgroup analysis and investigation of heterogeneity

 

1. Subgroup analyses

We did not anticipate any 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 were correct, we visually inspected the graph and successively removed studies outside of the company of the rest to see whether homogeneity was restored. For this review, we decided that should this occur with data contributing to the summary finding of no more than around 10% of total weighting, we would present the data. If this did not occur, we did not pool data and discuss issues. We know of no supporting research for this 10% cut-off, but we use prediction intervals as an alternative to this unsatisfactory state.

When unanticipated clinical or methodological heterogeneity was obvious, we simply stated hypotheses regarding this for future reviews or future versions of this review.

 

Sensitivity analysis

 

1. Implication of randomisation

We included trials in a sensitivity analysis if they did not state they were randomised but were described in some way as to imply randomisation. For primary outcomes, if no substantive difference was noted when the implied randomised studies were added to those with better description of randomisation, we entered all data from these studies.

 

2. Assumptions for lost binary data

When assumptions had to be made regarding people lost to follow-up and missing SD data (see Dealing with missing data), we compared the findings on primary outcomes when we used our assumption compared with completer data only. We undertook a sensitivity analysis to test how prone results were to change when 'completer' data only were compared with imputed data using the above assumption. If a substantial difference was noted, we reported results and discussed them but continued to employ our assumption.

 

3. Risk of bias

We analysed the effects of excluding trials that were judged to be at high risk of bias across one or more of the domains of randomisation (implied as randomised with no further details available): allocation concealment, blinding and outcome reporting for the meta-analysis of the primary outcome. If exclusion of trials at high risk of bias did not substantially alter the direction of effect or the precision of effect estimates, we included data from these trials in the analysis.

 

4. Imputed values

Had we imputed values, we would have also undertaken a sensitivity analysis to assess the effects of including data from trials for which we used imputed values for ICC in calculating the design effect in cluster randomised trials.

If imputed values are used in future versions of this review, we will note substantial differences in the direction or precision of effect estimates in any of the sensitivity analyses listed above, and we will not pool data from excluded trials with those of other trials contributing to the outcome but will present them separately.

 

5. Fixed-effect and random-effects models

We synthesised data using a fixed-effect model; however, we also synthesised data for primary outcomes using a random-effects model to evaluate whether this alters the significance of the results.

 

Results

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms
 

Description of studies

For a more detailed description of studies please see Characteristics of included studies and Characteristics of excluded studies.

 

Results of the search

The initial 2002 search yielded 114 electronic records. Of these, 79 were obtained as full publications. This initial selection process was independently repeated on a random sample of 12 records by TM, who achieved a Cohen's kappa of 0.63. Independent repeating of the assessment of the full publications on a 10% sample yielded 100% agreement. A further seven publications were obtained for perusal after handsearching of the references of obtained papers, but none of these could be included. High-yield journals were found to be those that had been handsearched previously by Cochrane Groups. An additional search by Janssen Medical Information Department led the review authors to obtain 24 further papers. Many of these had already been excluded. For the July 2005 update, the search yielded 51 citations (see Appendix 2) and a 2011 and a most recent 2013 search for the current update yielded 14 citations, of which three new studies were included. A total of 91 trials were assessed for eligibility; of the excluded publications, 60 were sufficiently close to the review's inclusion criteria to be included in the excluded studies table, and 31 were subsequently included in the data and analysis (see Figure 3 for study flow diagram). Contacting authors of included studies yielded further data (see Acknowledgements).

 FigureFigure 3. Study flow diagram: combined from the 2002 initial review search; the 2005 first update search; and the current 2013 update search.

 

Included studies

We included 32 studies published between 1971 and 2012. Most studies were described as 'randomised'; four trials did not make explicit reference to a randomised design; however, each employed a double-blind method (Andersen 1972; Kline 1977; Abuzzahab 1980*; Haas 1982). These trials were rated as 'high' risk of bias and were subject to a sensitivity analysis when data for primary outcomes of interest were provided. Only four trials provided an adequate description of randomisation (Pinard 1972; Clark 1975*; McInnes 1978; Barnes 1983); the remaining studies mentioned 'randomisation' and a double-blind technique but delivered no further information regarding the methods employed.

Six trials that were previously included in older versions of this review have been excluded; this decision was based on unbiased judgements made by the current review authors and was due to inadequate randomisation and blinding methods (Masiak 1976; McCreadie 1978; McCreadie 1983), or it occurred when closer inspection revealed misinterpreted data resulting when no overall extractable data (Sims 1975; Lapierre 1976) or no pre-cross-over data were available from a study with a cross-over design (Morris 1970).

 

1. Length of trials

Studies varied greatly in duration.

 

2. Participants

A total of 1277 participants were enrolled in the 32 included studies. Participants in most (17) of the studies appear to have been inpatients (Chouinard 1970; Huber 1971; Andersen 1972; Kudo 1972; Pinard 1972; Denijs 1973; Gowardman 1973; Gross 1974*; McInnes 1978; McCreadie 1980; Pecknold 1980; Chouinard 1982; Haas 1982; McCreadie 1982*; Silverstone 1984; McCreadie 1987; De Ronchi 1996; ), and participants in 13 of the studies appear to have been outpatients (Claghorn 1974; Kolivakis 1974; Clark 1975*; Amin 1977; Donlon 1977; Kline 1977; Vergara 1977; Falloon 1978; Abuzzahab 1980*; Wilson 1982*; Barnes 1983; Nishikawa 1985; Gunduz-Bruce 2013 ). Participants in Anumonye 1976 were initially hospitalised but could move to the community, and participants in Friedman 2011 were in either an inpatient or an outpatient setting.

Most studies included participants diagnosed with schizophrenia; however, the diagnostic criteria used were not specified in 63% of included studies, demonstrating what appear to be pragmatic or clinical diagnoses.

Not all included participants in McInnes 1978 had schizophrenia, as 15 were diagnosed as having schizophrenia; 2 had Korsakoff's and 1 had bipolar disorder. Further, in Andersen 1972, unclear diagnostic criteria were used to diagnose schizophrenia or 'paranoid syndrome'. These studies were included because most randomly assigned participants had schizophrenia. Two studies specifically selected people with severe negative symptoms (Kudo 1972; Sims 1975). Anumonye 1976 selected people with schizophrenia who also displayed impulsive, aggressive behaviour. Four studies focused on newly admitted, acutely ill people (Pecknold 1980; Chouinard 1982; Haas 1982; Silverstone 1984), whereas participants in Falloon 1978 were at the stage of being discharged after an acute episode. McCreadie 1987 randomly assigned people in their first episode of illness. People with schizophrenia complicated by substance misuse difficulties appear to have been frequently excluded by the study selection criteria. Ages ranged from 16 to 80 years, and the average age was 43 years. The sexes appeared evenly distributed overall, with 693 males and 584 females included.

 

3. Setting

Most included studies that provided details of where the trials took place were undertaken in the UK; study sites included a Community Nurse's Clinic in the London Borough of Southwark (Barnes 1983) and hospitals in Stone (Falloon 1978) and Scotland (McCreadie 1980; McCreadie 1982*; McCreadie 1987). Six studies took place in the USA, in a rehabilitative 'half-way-house' (Gross 1974*); in mental health research centres in Texas and Oklahoma (Claghorn 1974; Clark 1975*); in psychiatric hospitals (Wilson 1982*; Friedman 2011) and in an outpatient clinic in West Haven, Connecticut (Gunduz-Bruce 2013). Four further studies were undertaken in psychiatric hospitals in Canada three in Montréal (Pinard 1972; Pecknold 1980; Chouinard 1982) and one in Quebec (Chouinard 1970). Furthermore, two studies were undertaken in hospitals in Japan (Nishikawa 1985) and The Netherlands (Denijs 1973); another in a 'locked-ward' in New Zealand (McInnes 1978) and another in a state psychiatric hospital in Nigeria (Anumonye 1976). Huber 1971 was described as having taken place in a 'special research ward area', and De Ronchi 1996 took place at the Institute of Psychiatry of the University of Bologna, Italy. The remaining included studies provided no further details regarding settings and locations of the trials.

 

4. Study size

The overall sample size in each included study was generally small; Nishikawa 1985 was the largest study, randomly assigning 106 people, whilst McInnes 1978 was the smallest, with only 18 people randomly assigned. Most studies randomly assigned 20 to 45 people.

 

5. Interventions

The mean average dose of pimozide, based on the 22 studies that clearly reported it, was 11.52 mg/d, and the range, taken from 16 studies, was 1 to 70 mg/d. The highest average dose was 45.7 mg; however, the frequency was unclear (Huber 1971). The highest reported mean daily dose was 30 mg/d (Chouinard 1982). For comparison, the current recommended maximum dose of pimozide in the UK is 20 mg/d (BNF 2012). Twelve studies reported a maximum dose matching or exceeding this 20 mg limit (Huber 1971; Kolivakis 1974; Clark 1975*; Donlon 1977; Abuzzahab 1980*; McCreadie 1980; Pecknold 1980; Chouinard 1982; Haas 1982; McCreadie 1982*; Wilson 1982*; Silverstone 1984).

 
5.1 Pimozide versus placebo

Five studies compared pimozide with placebo, using low to medium doses of pimozide only one included study adopted a very high dose of up to 40 mg/d (Huber 1971). Most studies had more than two treatment arms and also present data for the comparison of 'pimozide versus any antipsychotic' (see later). Clark 1975* had three treatment arms and compared 2 to 16 mg/d of oral pimozide (average daily dose of 5.5 mg) with placebo (mean of 3 capsules/d) or 75 to 375 mg/d thioridazine. In the trial conducted by Falloon 1978, participants received oral pimozide at a maximum dose of 16 mg/d(median 8 mg/d) plus inert fluphenazine injections, or they received active fluphenazine injections at a maximum dose of 50 mg/wk (median 25 mg/2 wk) plus inert pimozide tablets. The pimozide/placebo comparison in this study was made during a six-month follow-up period in which all injections (both active and placebo) were withdrawn from a selection of participants who had remained well throughout the previous 12 months of the trial. Participants in the Gross 1974* study were assigned to receive one of three treatments: 2 to 12 mg/d (mean 6.3 mg/d) pimozide, placebo or trifluoperazine (see later for the trifluoperazine comparison); and Pinard 1972 had five treatment arms, in which participants received 3 mg/d pimozide; 6 mg/d pimozide; placebo; or 5 mg/3 times daily or 15 mg/d trifluoperazine (see later). The results for these groups were combined to equal three treatment arms for the purposes of presenting data in the meta-analysis (i.e., data from the two pimozide groups were presented together, as were the data from the two trifluoperazine groups). Huber 1971 was the only study to compare pimozide versus placebo with no additional treatment arms, in which participants received a high dose of up to 40 mg/d oral pimozide or an oral placebo. Finally, Denijs 1973 were employed in a two-month maintenance phase when all participants received a mean of 7.72 mg/d pimozide. Those who had not relapsed during this period were then randomly assigned to continue pimozide or receive placebo (we have presented the data from this withdrawal study separately).

 
5.2 Pimozide versus any antipsychotic

Twenty-six studies compared pimozide with antipsychotic drugs. Four studies compared pimozide with a long-acting depot preparation, fluphenazine decanoate (Falloon 1978; McCreadie 1980; McCreadie 1982*; Barnes 1983). Among these studies, one described the use of a moderate pimozide dose of maximum 16 mg (Falloon 1978); another used a low mean dose of 5.3 mg/d pimozide (Barnes 1983) and the final two studies by the same author employed an intermittent regimen of pimozide, with a maximum 32 mg every four days a week (McCreadie 1980) or 40 mg once a week (McCreadie 1982*). Dosages of fluphenazine decanoate were moderate and comparable amongst the studies taken intermittently, these ranged from a mean of 8.2 mg (Barnes 1983) or 12.5 mg a week (McCreadie 1980) to a median of 25 mg every two weeks (Falloon 1978) or a mean of 14 mg biweekly (McCreadie 1982*). Three studies compared a moderate to high range of 10 to 20 mg/d pimozide with a moderate mean dose of oral fluphenazine (Chouinard 1970; Donlon 1977; Abuzzahab 1980*).

We found seven studies that compared pimozide with trifluoperazine both Amin 1977 and Vergara 1977 compared low to moderate range pimozide of 2 to 12 mg daily with low to moderate range trifluoperazine 5 to 30 mg daily (dosages in Vergara 1977 were adjusted on the basis of clinical criteria to achieve 'optimal therapeutic effect'). Participants in Andersen 1972 received low to moderate range pimozide of 3 to 14 mg daily versus a lower dose range of 3 to 12 mg trifluoperazine (mean dose 4.34 mg twice daily); Claghorn 1974 participants received a low mean dose of 5.16 mg a day pimozide versus a low dose of 12.53 mg trifluoperazine daily; people in Gross 1974* received a moderate dose of pimozide, with a mean of 6.3 mg a day (range 2 to 12 mg daily) versus a moderate dose of trifluoperazine, with a mean 17.5 mg daily (range 5 to 30 mg). Kline 1977 compared the effects of pimozide in the range of 4 to 12 mg/d (however, reporting a mean dose of 15.7 mg/d) with trifluoperazine in the range 5 to 25 mg daily (mean 6.5 mg). Finally, Pinard 1972, as discussed earlier, had five treatment arms and compared the lower doses of pimozide with 5 mg trifluoperazine three times daily or 15 mg trifluoperazine daily these results were combined in the meta-analysis.

Five studies compared pimozide with chlorpromazine Anumonye 1976 compared a low dose of pimozide (mean 1 to 4 mg/d) with a moderate dose of 300 to 600 mg chlorpromazine a day, and participants in Kolivakis 1974 received a moderate mean daily dose of 7 mg pimozide versus a lower dose of 75 to 450 mg chlorpromazine (mean 216 mg/d). The other three studies employed higher doses of each drug: In Chouinard 1982, participants received a high mean dose of 30 mg pimozide a day (range 10 to 70 mg) versus 300 to 1500 mg chlorpromazine a day; participants in Pecknold 1980 received pimozide in the range of 10 to 70 mg/d versus 300 to 2100 mg chlorpromazine a day; and Wilson 1982* participants received a maximum dose of 20 mg pimozide per day, with a mean dose of 7.3 mg versus a maximum of 950 mg chlorpromazine (moderate mean dose 381 mg/d).

Three studies compared pimozide with haloperidol: A low mean dose of 5.1 mg pimozide was compared with a moderate maximum dose of 14 mg/d of haloperidol in Gowardman 1973; higher doses of pimozide (40 to 60 mg/ d) were compared with high-dose haloperidol (40 to 60 mg/d) in Haas 1982, while participants in Silverstone 1984 received 5 to 50 mg a day of pimozide or haloperidol.

Higher doses of pimozide (mean 18.8 mg/d) were compared with high-dose flupenthixol (mean 20 mg/d) in McCreadie 1987, and low-dose pimozide was compared with thioridazine in two studies at a range of 75 to 375 mg daily (Clark 1975*) or 25 mg or 75 mg daily (Nishikawa 1985;this study also looked at the effects of combined pimozide and thioridazine; see later).

We found only one study that compared pimozide (3 to 8 mg/d) plus placebo with atypical antipsychotic drugs. Kudo 1972 used carpipramine plus placebo as the comparator, which has a similar biochemical profile but a shorter half-life than pimozide.

 
5.3 Pimozide + any antipsychotic versus any antipsychotic

One study made this comparison while using low to medium doses of pimozide. Participants in Nishikawa 1985 received, orally, 2 mg pimozide augmented with 25 mg or 75 mg thioridazine per day, or 6 mg pimozide augmented with 25 mg or 75 mg a day versus 25 mg or 75 mg thioridazine alone. Furthermore, each drug was combined with 10 mg nitrazepam for insomnia and 2 mg trihexyphenidyl for drug-induced Parkinsonism.

 
5.4 Pimozide + any antipsychotic versus any antipsychotic + placebo

In the Friedman 2011 study, participants received 6.48 mg/d (mean) pimozide augmented with 518.8 mg/d (mean) clozapine, or clozapine 478.1 mg/d (mean) with placebo. Participants in Gunduz-Bruce 2013 received oral pimozide starting at 1 mg/d and maximum 4 mg/d by the end of week 7 augmented with clozapine (which was a stable dose determined prerandomisation of at least 350 ng/mL targeted), or they received clozapine (again, at a targeted 350 ng/mL) with placebo sugar pill.

 
5.5 Pimozide + any antipsychotic versus any antipsychotic + any antipsychotic

Only one study made this comparison; an atypical antipsychotic was compared with pimozide in De Ronchi 1996, which compared 4 mg/d pimozide plus 2 mg/d haloperidol with 200 mg/d levosulpiride plus 2 mg/d haloperidol.

 

6. Outcomes

Most outcomes were reported as dichotomous (yes-no/binary outcomes) and are presented in this review as such. Scale-derived data on global state (Clinical Global Impression (CGI)) were usually categorical and therefore easily dichotomised. This process of dividing categorical data into two is vulnerable to the inclusion of bias, that is, someone seeing the data might conceivably choose a point of division in a biased fashion. The 'cut-off' used in this review was chosen on the basis of advice given blind to the data. Details of scales that provided usable data are shown below. We have given reasons for exclusion of data from instruments under 'Outcomes' in the 'Excluded studies' section.

No study reported on negative symptoms as an outcome. Neither were there usable cognitive outcomes. None of the included studies attempted to quantify levels of satisfaction or to provide any direct economic evaluation of pimozide.

 

7. Outcomes: scales providing usable data

 
7.1 Global state

7.1.1 Clinical Global Impression CGI (Guy 1976)
This three-item observer-rated scale measures severity of illness (CGI-S), global improvement (CGI-I) or change (CGI-C) and therapeutic response, which are rated on a 7-point scale from 1 = normal to 7 = extremely unwell; lower scores are indicative of decreased severity and/or greater recovery, and the treatment response ratings, which take into account both therapeutic efficacy and treatment-related adverse effects, range from 0 = marked improvement and no side effects, to 4 = unchanged or worse, with side effects outweighing therapeutic effects. Each component of the CGI is rated separately, and the instrument does not provide a global score. Pecknold 1980; De Ronchi 1996; Friedman 2011 and Gunduz-Bruce 2013 are the only studies that have reported data using this scale.

Other studies provided binary data using this scale, including Gross 1974*; Clark 1975*; Amin 1977; Vergara 1977 and De Ronchi 1996; andwhich measured numbers of participants as 'improved' or 'much improved' on the basis of CGI ratings. Haas 1982 measured improvement/no improvement using the CGI by including those still rated as 'severely' or 'moderately ill' by 30 days. Huber 1971; Kolivakis 1974 and Kline 1977 adopted a similar approach by assessing psychiatrists' CGI ratings; participants who were still rated as 'not changed', 'minimally worse', 'much worse' or 'very much worse' were considered not improved.

 
7.2 Mental state

7.2.1 Brief Psychiatric Rating Scale BPRS (Overall 1962)
The BPRS is an 18-item scale that measures positive symptoms, general psychopathology and affective symptoms. The original scale consists of 16 items, but the revised 18-item scale is more commonly used. Scores range from 0 to 126. Each item is rated on a 7-point scale, with 0 = 'not present' to 7 = 'extremely severe' and with high scores indicating more severe symptoms. Gunduz-Bruce 2013 and Pecknold 1980 were the only studies to report data from this scale. Anumonye 1976 and Kline 1977 provided binary data using the BPRS to measure improvement/no improvement, including results from participants who had achieved an 'absent' level of improvement (Anumonye 1976) and from participants with either 'no change' or a 'worsened' level of improvement (Kline 1977).

7.2.2 Montgomery Asberg Depression Rating Scale MADRS (Montgomery 1979)
This scale is used to assess the severity of a range of psychotic symptoms in people with schizophrenia. It is a 65-item comprehensive psychopathology scale that is used to identify the 17 most commonly occurring symptoms in primary depressive illness. Ratings are based on 10 items. A high score indicates more severe and/or extensive symptoms. Silverstone 1984 was the only study to report data from this scale.

7.2.3 Positive and Negative Syndrome Scale PANSS (Kay 1987)
PANSS was developed from the BPRS and the Psychopathology Rating Scale. It is used to evaluate positive, negative and other symptom dimensions in schizophrenia. The scale consists of 30 items, each measured on a 7-point scoring system ranging from 1 = absent to 7 = extreme. Only Friedman 2011 provided data using this scale.

7.2.4 Scale for the Assessment of Negative Symptoms SANS (Andreasen 1982)
The SANS measures the incidence and severity of negative symptoms using a 25-item scale and a 6-point scoring system, with 0 = better to 5 = worse; a higher score indicates a more severe experience of negative symptoms. De Ronchi 1996 and Gunduz-Bruce 2013 were the only studies to provide data using this scale.

7.2.5 Scale for the Assessment of Positive Symptoms SAPS (Andreasen 1982)
The SAPS, which is intended for use as a complementary instrument to the SANS, measures positive symptoms using a 34-item scale with a similar 6-point scoring system, by which 0 = better to 5 = worse. The 34 items are split into four sections, each measuring specific positive symptoms, including hallucinations, delusions, bizarre behaviour and positive formal thought disorder. De Ronchi 1996 was the only study that provided usable data from this scale.

 
7.3 Social functioning

7.3.1 Social Behaviour Assessment Schedule SBAS (Platt 1980)
The SBAS is a standardised, semi-structured interview with the participant's closely involved family or friend (the 'informant'), undertaken without the participant present. The scale takes around 60 to 90 minutes to complete and measures both objective and subjective elements, referred to as the 'objective burden' (the effect of the participant's behaviour on others) and the 'subjective burden' (the distress directly inflicted by the participant upon others). The objective burden is assessed in a separate section, which takes into account changes that have occurred in the household and in the lives of the informant and relatives; the subjective burden is rated separately for each item in sections B (behaviour, onset, distress), C (change in social performance, onset of change, distress) and D (effects on children, onset, patient-relatedness, distress, effects on informant, effects on others). Section E assesses concurrent events of relevance to the participant's current situation (again, measuring onset and independence), and Section F assesses the amount of help and support that the informant receives from friends/relatives (support, informant's housing situation). Ratings are assigned for each item from a 0, 1, 2, 3 rating schema, where a lower score indicates a better outcome. Barnes 1983 was the only study to employ this scale, which provided skewed data.

 
7.4 Quality of life

7.4.1 Specific Level of Functioning SLOF (Schneider 1983)
The SLOF is a clinician-based rating scale that consists of 43 items, all of which are grouped into six behavioural and functional areas, including social skills, problem behaviours, self-care deficits, community-living skills, vocational functioning and interpersonal relationships. Each item is judged on a 5-point Linkert-type scale from 0 to 4, with a high score indicating a better level of functioning. Friedman 2011 was the only study that provided data using this scale; however, these are skewed.

 
7.5 Adverse effects

7.5.1 Extrapyramidal Symptom Rating Scale ESRS (Chouinard 1980)
The ESRS is a clinician-rated scale that assesses four types of drug-induced movement disorders, including Parkinsonism, akathisia, dystonia and tardive dyskinesia. Seven items are included in the first section of the scale the 'questionnaire' and are judged on a 4-point scale of 0 to 3, where 0 = absent and 3 = severe. The second section the 'examination' examines three items, including tremor, bradykinesia, and gait and posture. These are rated for individual body parts on a two-dimensional continuum of severity and frequency. Friedman 2011 was the only study that provided usable data from this scale; these data were skewed.

7.5.2 Extrapyramidal Side Effects Scale EPS (Simpson 1970)
This observer-rated scale is designed to assess parkinsonian and related extrapyramidal side effects. The scale consists of ten items, including gait, arm dropping, shoulder shaking, elbow rigidity, wrist rigidity, leg pendulousness, head dropping, glabella tap, tremor and salivation; each is measured on a 5-point scoring system from 0 = absent to 4 = severe. Pecknold 1980 was the only study that reported data using this scale; however, these were, again, skewed.

 

Excluded studies

We excluded 60 studies. Most of these were reviews, which did not provide usable data, or were not randomised studies. Two trials did not provide the required intervention (van Kammen 1982; Ibarra 1996), and two did not focus on people with psychotic illness (Reyntjens 1972; Poldinger 1976). One multi-centre study (Brugmans 1968) would have been included if it had presented data by experimental group. Both Crow 1986 and Johnstone 1997 compared several different antipsychotic drugs, including pimozide, with placebo but did not report data by individual drug. Five small trials used cross-over methodology and did not present data derived from individual stages of the study (Morris 1970; Kenway 1971; Svestka 1972; Van Wyck 1972; Smythies 1974). We are seeking data from these studies for subsequent versions of this review.

 

Awaiting assessment

One study awaits translation (Umene 1972).

 

Ongoing studies

We know of no ongoing studies.

 

Risk of bias in included studies

For a graphical overview, please see Figure 4 and Figure 5.

 FigureFigure 4. Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
 FigureFigure 5. Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

 

Allocation

Twenty-eight studies were reported to be randomised; four studies were rated as 'high risk' of bias because no description of randomisation was provided (Andersen 1972; Kline 1977; Abuzzahab 1980*; Haas 1982); only 'double blind allocation' had taken place. Inclusion of data from these studies did not substantially alter most results of the analyses. Therefore we decided to include these studies in the overall analyses. When these studies appeared in our primary outcomes, they were subject to a sensitivity analysis this has been provided in the data and analysis sections (sensitivity analysis:  Analysis 7.1;  Analysis 7.2). Only four studies provided adequate information on randomisation techniques (Pinard 1972; Clark 1975*; McInnes 1978; Barnes 1983). The remaining 24 studies were rated as 'unclear' risk of bias, as they had mentioned that randomisation had been employed but did not provide any further details.

Eighteen of the 28 studies reported as randomised included equal numbers of participants in the experimental and control groups without stating that they had used a matched pair design. The method of randomisation generally was not described with the exceptions of McInnes 1978, who reported 'flicking a coin', and Pinard 1972, who used a 'random number table'. Allocation concealment was generally poorly reported. Exceptions were Gowardman 1973, who reported allocation by the hospital pharmacist, suggesting adequate separation of allocation and treatment, and McInnes 1978, who employed a 'double blind code'. Donlon 1977 reported allocation by a 'disinterested third party'. Poor reporting of allocation concealment has been associated with an overestimation of effect of about 30% to 40% (Schulz 1994; Moher 1998), which is likely with the remaining trials.

 

Blinding

Overall description of blinding was poor throughout the records of the included studies; studies were described as 'double blind' with two exceptions, where Anumonye 1976 and De Ronchi 1996 used a single-blind method. Eighteen studies briefly mentioned precautions that they used to maintain blindness (Chouinard 1970; Andersen 1972; Kudo 1972; Denijs 1973; Gowardman 1973; Claghorn 1974; Gross 1974*; Kolivakis 1974; Clark 1975*; Falloon 1978; McCreadie 1980; Chouinard 1982; Haas 1982; McCreadie 1982*; Wilson 1982*; Barnes 1983; Silverstone 1984; Nishikawa 1985). Twelve studies reported identical appearing drug treatment; Nishikawa 1985, in addition, reported that taste was made identical by adding a "common gastric acid". Seven studies reported adding oral placebo to the experimental regimens to ensure that all participants had the same dosing frequency, and another three reported giving drug treatment as a single daily dose to all participants. Six studies that compared pimozide with a depot antipsychotic injection reported using a 'double dummy design' (Andersen 1972; Falloon 1978; McCreadie 1980; McCreadie 1982*; Barnes 1983; McCreadie 1987), meaning that they had given either placebo injection or oral placebo to trial participants.

Blindness was not reported to have been tested in any of the included studies. Two questions: one to the participant "What do you think you have been given?" and one to the rater "What drug do you think this person was allocated?" would have clarified the situation. Failure to test double blinding leaves doubt as to how much observation bias is included in the data.

 

Incomplete outcome data

Only three studies reported data with 100% follow-up of participants (Gowardman 1973; Anumonye 1976; Chouinard 1982), and LOCF was used in six of the included studies, in which loss was reported at 35% or less (Chouinard 1970; Denijs 1973; Amin 1977; Barnes 1983; De Ronchi 1996; Friedman 2011). One study reported a large number of participants lost to follow-up (Abuzzahab 1980*), with only 24% of participants completing the full three years of the study only the outcome of 'leaving the study early' was presented in the data and analysis as per our protocol regarding Dealing with missing data.

Overall, poor reporting at each stage of the included studies made it difficult to ascertain the true numbers of participants who left the study early use of a flow chart would have been an appropriate measure to accurately reflect these numbers. Seventeen studies provided details of participants leaving the study early (Chouinard 1970; Huber 1971; Andersen 1972; Denijs 1973; Kolivakis 1974; Clark 1975*; Anumonye 1976; Amin 1977; Donlon 1977; McCreadie 1980; Pecknold 1980; Haas 1982; McCreadie 1982*; Wilson 1982*; Silverstone 1984; McCreadie 1987; De Ronchi 1996 ), and five studies did not make clear the numbers of participants lost to follow-up (Pinard 1972; Claghorn 1974; Kline 1977; McInnes 1978; Nishikawa 1985). Most studies did not indicate whether they reported completer-only data; did not specify whether completer-only or ITT analyses were employed; or presented highly skewed data.

 

Selective reporting

Most studies were rated as high risk of bias, as not all outcomes were reported, in particular, data from rating scales went unreported from the time intervals specified in the study protocols. Continuous scale data were often poorly reported and either lacked explicit statements regarding the denominator and standard deviation or were presented only as significance tests.

Six of the 21 studies providing global outcomes appear to have presented their categorical global state data using a 'last observation carried forward' approach for those who left the study (Chouinard 1970; Denijs 1973; Amin 1977; Barnes 1983; De Ronchi 1996; Friedman 2011; Gunduz-Bruce 2013). Sometimes this was stated in the text, but in other instances, it was apparent only from the tables. It is difficult to know how this practice affects results in this review, but it might (Cotton 2006; Leucht 2007). We have labelled these data as 'LOCF'.

Ten studies reported relapse as an outcome, but none made the criteria explicit (Chouinard 1970; Andersen 1972; Denijs 1973; Amin 1977; Vergara 1977; Falloon 1978; McCreadie 1980; McCreadie 1982*; Barnes 1983; Nishikawa 1985); criteria appear to have been pragmatic clinical diagnoses that included 'increase in/exacerbation of positive symptoms' (Denijs 1973; McCreadie 1980; McCreadie 1982*) and 'clinical deterioration' (Vergara 1977). Relapse data therefore were labelled 'clinically diagnosed' rather than 'defined by'.

 

Other potential sources of bias

 
1. Funding

Twenty-one of the included studies provided information regarding funding sources; of these, three did not specify any involvement with pharmaceutical companies, instead receiving support grants from the United States Public Health Service and the Psychopharmacology Fund, Minneapolis, Minnesota (USA) (Abuzzahab 1980*); a grant from the LEO Research Foundation, Helsingborg (Sweden) (Andersen 1972) or support from the Stanley Medical Research Institute, Chevy Chase, Maryland (USA) (Gunduz-Bruce 2013). The remaining studies were supplied grants as well as financial or other support from pharmaceutical companies: each of eighteen studies reported receipt of funding, advice, support and drugs provided by specific pharmaceutical companies. This included funding/grants or study drugs supplied by McNeil Laboratories, Fort Washington, Pennsylvania (USA) (Chouinard 1970; Huber 1971; Gross 1974*; Kolivakis 1974; Clark 1975*; Donlon 1977; Kline 1977; Pecknold 1980; Wilson 1982*) and drugs supplied by Ravizza Farmaceuti, Milan (Italy) (De Ronchi 1996). Seven of the included studies received study drugs, advice or financial or other support from the pharmaceutical company responsible for developing pimozide, Janssen Pharmaceutica (Belgium) (Denijs 1973; Falloon 1978; McCreadie 1980; Haas 1982; McCreadie 1982*; Silverstone 1984; McCreadie 1987). One author in the Silverstone 1984 study was an employee of Janssen Pharmaceutica at the time of the trial. Further, Friedman 2011 received funding from grant MH0678060182 of the National Institute of Mental Health (NIMH) but reported an extensive list of study authors' commercial interests, including previous grant support and contacts of employment (see Characteristics of included studies for further details).

 
2. Rating scales

Most of the included studies did not indicate whether raters of measurement scales were independent of treatment; only seven studies reported using independent assessors (see Characteristics of included studies). Clark 1975* reported that the raters were a "research nurse and project psychiatrist". The research nurse data were thought more likely to be independent and are presented in this review, although the project psychiatrist data are very similar. Because the scale-rated outcomes were largely rated by a person who was unlikely to be disinterested in the final result, most of these are presented in this review with the warning, 'prone to bias'. Only seven studies gave details on how rating scales were administered and the extent to which raters were independent (Andersen 1972; Pinard 1972; Anumonye 1976; Falloon 1978; McCreadie 1980; McCreadie 1982*; Friedman 2011).

 

Effects of interventions

See:  Summary of findings for the main comparison PIMOZIDE versus PLACEBO for schizophrenia or related psychoses;  Summary of findings 2 PIMOZIDE versus ANY ANTIPSYCHOTIC for schizophrenia or related psychoses;  Summary of findings 3 PIMOZIDE + ANY ANTIPSYCHOTIC versus ANY ANTIPSYCHOTIC for schizophrenia or related psychoses;  Summary of findings 4 PIMOZIDE + ANY ANTIPSYCHOTIC versus ANTIPSYCHOTIC + PLACEBO for schizophrenia or related psychoses;  Summary of findings 5 PIMOZIDE + ANY ANTIPSYCHOTIC versus ANTIPSYCHOTIC + ANTIPSYCHOTIC for schizophrenia or related psychoses

 

1. COMPARISON 1: PIMOZIDE versus PLACEBO

Five studies were included in this comparison, with a total of 163 participants. Only two studies provided data for the primary outcomes of mental state and extrapyramidal symptoms, which included specific symptoms (n = 45; Huber 1971; Clark 1975*). Each of the remaining studies provided data for the secondary outcomes, including global state, adverse effects and leaving the study early all data presented are binary, and no continuous data are available for analysis.

 

1.1 Global state

 
1.1.1 Relapse

Overall results were equivocal for global state, with no real differences noted between those receiving pimozide and those receiving placebo. Only one small study provided data for relapse, which by medium term slightly favoured pimozide, indicating a greater risk of relapse for those receiving placebo. However, this was not significant (1 RCT n = 20, RR 0.22 CI 0.03 to 1.78;  Analysis 1.1).

 
1.1.2 No improvement

Pimozide was slightly favoured by both short (2 RCTs n = 40, RR 0.61 CI 0.15 to 2.45) and medium term (2 RCTs n = 62, RR 0.84 CI 0.67 to 1.06;  Analysis 1.2) with greater risk of relapse noted for people receiving placebo; again, this was not significant. Results by short term demonstrate a moderate level of heterogeneity (P = 0.15, I² = 53%) and are presented using a random-effects model (no significant difference between the results is evident if a fixed-effect model is used).

 

1.2 Mental state

 
1.2.1 Specific symptoms

Again, results for mental state show no effect of pimozide. Slight, non-significant favour was noted with placebo regarding anxiety and tension experienced by people receiving pimozide by medium term ( 1 RCT n = 25, RR 2.06 CI 0.09 to 46.11) and auditory hallucinations by short term (1 RCT n = 19, RR 3.30 CI 0.15 to 72.08). However, slight favour was seen with pimozide for the incidence of depression, with greater risk observed by medium term for those receiving placebo (1 RCT n = 25, RR 0.67 CI 0.05 to 9.47;  Analysis 1.3); again, this was not significant.

 

1.3 Adverse effects

Data are very limited for all adverse effect outcomes, with only three studies presenting usable data (total n = 111); none of the data are pooled.

 
1.3.1 Extrapyramidal adverse effects specific

Only two studies presented data for this outcome. Huber 1971 presented short-term data and found no difference between groups for akathisia (1 RCT n = 19, RR 5.50 CI 0.30 to 101.28), rigidity (1 RCT n = 19, RR 5.50 CI 0.30 to 101.28) or tremor (1 RCT n = 19, RR 3.30 CI 0.15 to 72.08). Similar equivocal results were found by Clark 1975*, who presented medium-term data for dystonia (1 RCT n = 25, RR 0.67 CI 0.05 to 9.47), akathisia (1 RCT n = 25, RR 0.22 CI 0.03 to 1.85), rigidity (1 RCT n = 25, RR 1.33 CI 0.14 to 12.82), tremor (1 RCT n = 25, RR 1.0 CI 0.20 to 4.95) and feeling restless (1 RCT n = 25, RR 2.00 CI 0.24 to 16.61;  Analysis 1.4).

 
1.3.2 Anticholingenic effects

It was found that pimozide was no more likely than placebo to cause blurred vision (1 RCT n = 25, RR 1.00 CI 0.20 to 4.95) or dry mouth (1 RCT n = 25, RR 0.23 CI 0.01 to 5.12;  Analysis 1.5).

 
1.3.3 Cardiovascular effects

No clear difference between pimozide and placebo was found over a range of cardiovascular outcomes. Huber 1971 presented identical data for two outcomes: increase in blood pressure and electrocardiographic (ECG) changes. The difference between groups was not statistically significant for either outcome (1 RCT n = 19, RR 3.30 CI 0.15 to 72.08). Clark 1975* found that pimozide did not produce more severe dizziness than was seen with placebo (1 RCT n = 25, RR 0.67 CI 0.11 to 3.99;  Analysis 1.6), nor did it produce increased chest pain.

 
1.3.4 Abnormal laboratory tests

Again, individual studies with small sample sizes presented data for abnormal laboratory tests; tests were equivocal for each outcome with confidence intervals for best estimate of effect, including both no effect and appreciable benefit/ harm. Slight favour was noted with pimozide when any abnormal haematological data were assessed (1 RCT n = 25, RR 0.53 CI 0.19 to 1.51), when white cell count increased (1 RCT n = 19, RR 0.56 CI 0.06 to 5.14) and when blood urea nitrogen balance was abnormal (1 RCT n = 25, RR 0.23 CI 0.01 to 5.12); slight favour was seen with placebo with liver function abnormalities (1 RCT n = 25, RR 2.06 CI 0.09 to 46.11) and increased alkaline phosphatase (1 RCT n=19, RR 1.67 CI 0.36 to 7.82;  Analysis 1.7). These findings were not significant.

 
1.3.5 Central nervous system effects

Small studies again presented data for this outcome; no differences were found between groups for headaches (1 RCT n = 25, RR 2.33 CI 0.60 to 9.02) or sedation by medium term (1 RCT n = 25, RR 0.33 CI 0.03 to 3.20), or for drowsiness by short term (1 RCT n = 19, RR 0.22 CI 0.01 to 4.05). It did appear as though pimozide had a beneficial effect on insomnia, but the result just failed to reach significance (1 RCT n = 25, RR 0.58 CI 0.31 to 1.09;  Analysis 1.8).

 
1.3.6 Dermatological effects

Data show that pimozide is no more likely than placebo to cause dermatitis (1 RCT n = 25, RR 0.33 CI 0.03 to 3.20;  Analysis 1.9).

 
1.3.7 Endocrine effects

No instances of galactorrhoea were reported in either group ( Analysis 1.10).

 
1.3.8 Gastrointestinal effects

Data show that pimozide was no more likely than placebo to cause acute cholecystitis (1 RCT n = 45, RR 1.41 CI 0.06 to 32.53). Pimozide was no more likely to cause diarrhoea (1 RCT n = 25, RR 0.22 CI 0.03 to 1.85) or nausea (1 RCT n = 25, RR 0.22 CI 0.03 to 1.85;  Analysis 1.11), and no instances of constipation were recorded in either group.

 
1.3.9 Other

Medium-term data from one small study reveal no difference between groups for anorgasmia (1 RCT n = 25, RR 2.06 CI 0.09 to 46.11). A higher risk of intensified symptoms was noted amongst people receiving placebo by short term (1 RCT n = 45, RR 0.09 CI 0.00 to 1.83) and by medium term (1 RCT n = 41, RR 0.42 CI 0.15 to 1.16); however, this was not significant. Pimozide was no more likely than placebo to cause weight gain (1 RCT n = 25, RR 3.44 CI 0.18 to 64.88) or weight loss by medium term (1 RCT n = 25, RR 0.44 CI 0.09 to 2.20). It was found that pimozide did not cause greater weight loss than placebo in the short term (1 RCT n = 19, RR 0.63; 95% CI 0.28 to 1.46). Data for ocular pigment deposit (Huber 1971) show that pimozide was no more likely than placebo to produce an increase or a decrease (1 RCT n = 10, RR 0.37 CI 0.02 to 8.01;  Analysis 1.8). These results were not significant. Individual studies presented data for nasal congestion (Clark 1975*), which demonstrated slight favour for pimozide andgreater risk for those receiving placebo (1 RCT n = 25, RR 0.23 CI 0.01 to 5.12). Nor were any differences in unpleasant dreams found between the groups (1 RCT n = 19, RR 3.30 CI 0.15 to 72.08;  Analysis 1.12).

 

1.4 Leaving the study early

 
1.4.1 Because of adverse effects

Adverse effects did cause some attrition, but no clear differences between groups were noted by short term (1 RCT n = 45, RR 0.23 CI 0.02 to 2.29) or by medium term (1 RCT n = 59, RR 0.19 CI 0.01 to 3.75;  Analysis 1.13).

 
1.4.2 For any reason

Three small studies (n = 90) reported data on attrition for any reason, and no difference between pimozide and placebo was seen at short term (3 RCTs n = 90, RR 1.01 CI 0.30 to 3.39); however, evidence suggested slight heterogeneity (P = 0.29, I² = 20%). Medium-term results did show a difference favouring pimozide this difference was just statistically significant (1 RCT n = 66, RR 0.35 CI 0.17 to 0.72;  Analysis 1.14).

 

2. COMPARISON 2: PIMOZIDE versus PLACEBO: WITHDRAWAL STUDY

Participants in Denijs 1973 (n = 40) were initially maintained on pimozide for two months. They were then allocated to pimozide continuation or to withdrawal onto placebo. This study, therefore, is reported separately. Data were reported for the primary outcome of mental state, as well as for secondary outcomes of global state, adverse effects and leaving the study early. All data presented are binary, and no continuous data are available for analysis.

 

2.1 Global state

 
2.1.1 Relapse clinical diagnoses

Slight, non-significant favour for pimozide was noted, with higher risk of relapse evident at medium term amongst participants receiving placebo (1 RCT n = 40, RR 0.33 CI 0.01 to 7.72;  Analysis 2.1).

 

2.2 Mental state

 
2.2.1 Use of additional medication

Significantly more people in the placebo group than in the pimozide group needed additional haloperidol (1 RCT n = 40, RR 0.18 CI 0.06 to 0.51;  Analysis 2.2).

 

2.3 Adverse effects

 
2.3.1 Death

One death in the pimozide group was reported as due to "toxi-infectious shock"; this did not produce a significant difference between groups (1 RCT n = 40, RR 3.00 CI 0.13 to 69.52;  Analysis 2.3).

 

2.4 Leaving the study early

 
2.4.1 Any reason

No difference was noted between groups for study attrition at six months for any reason (1 RCT n = 40, RR 1.00 CI 0.7 to 14.90;  Analysis 2.4).

 

3. COMPARISON 3: PIMOZIDE versus ANY ANTIPSYCHOTIC

Twenty-six of the included studies (n = 921) provided data for this comparison; of these, two studies provided data for the primary outcome of mental state: no improvement (Anumonye 1976; Kline 1977; n = 54); 12 RCTs provided data for extrapyramidal adverse effects (Chouinard 1970; Kudo 1972; Claghorn 1974; Clark 1975*; Amin 1977; Kline 1977; Falloon 1978; McCreadie 1980; Pecknold 1980; Haas 1982; McCreadie 1982*; Silverstone 1984; n = 412); four RCTs provided data for incidence of use of antiparkinson drugs (Andersen 1972; Kline 1977; McCreadie 1980; Haas 1982; n = 145) and one study provided data for average endpoint score using the EPS (Pecknold 1980; n = 16); however, these results were skewed. The remaining studies provided data for secondary outcomes, including service utilisation, global state, mental state, adverse effects, social functioning and leaving the study early. Most outcomes are dichotomous; however, some scale data were available in some studies (Pecknold 1980; Barnes 1983; Silverstone 1984).

 

3.1 Service utilisation

 
3.1.1 Hospital admission

No difference was reported between groups of numbers of participants admitted to hospital by medium term when pimozide was compared with fluphenazine decanoate (1 RCT n = 44, RR 0.83 CI 0.35 to 1.98;  Analysis 3.1).

 

3.2 Global state

 
3.2.1 Relapse clinical diagnoses

Two studies found no short-term difference between those taking pimozide and others allocated to two different antipsychotic drugs, including trifluoperazine and fluphenazine (2 RCTs n = 60, RR 1.67 CI 0.23 to 11.87). At medium term, again, no clear differences between groups were noted when pimozide was compared with trifluoperazine, fluphenazine decanoate or thioridazine (7 RCTs n = 227, RR 0.82 CI 0.57 to 1.17;  Analysis 3.2).

 
3.2.2 No improvement

In five studies, pimozide was slightly favoured over other oral antipsychotic drugs (including thioridazine, haloperidol, chlorpromazine and carpipramine) in the short term (5 RCTs n = 146, RR 0.75 CI 0.48 to 1.17), indicating a slightly greater risk of no improvement for people receiving other antipsychotics. Data for medium term were equivocal when pimozide was compared with trifluoperazine, thioridazine or chlorpromazine (6 RCTs n = 186, RR 0.99 CI 0.75 to 1.31;  Analysis 3.3), and a slight degree of heterogeneity was oted (P = 0.24, I² = 26%).

 
3.2.3 Average score (CGI)

In the one study that reported data using the CGI for global impression, no difference between groups was reported when pimozide was compared with chlorpromazine in the short term (1 RCT n = 16, MD -0.15 CI -1.25 to 0.95;  Analysis 3.4).

 

3.3 Mental state

 
3.3.1 Average score (BPRS)

In the short term, pimozide was slightly favoured over chlorpromazine in the one study that reported data using this scale; however, this finding was not significant (1 RCT n = 16, MD -13.88 CI -28.21 to 0.45;  Analysis 3.5).

 
3.3.2 Average score (MADRS), skewed

Only one study reported data using this outcome scale (n = 22); results are skewed; however, the mean scores indicate no differences between people receiving pimozide or haloperidol ( Analysis 3.6).

 
3.3.3 No improvement

The binary data, which were measured using the BPRS, demonstrated no apparent differences between people receiving pimozide versus chlorpromazine or trifluoperazine at short term (2 RCTs n = 54, RR 0.92 CI 0.43 to 1.96) and at medium term (1 RCT n = 23, RR 1.09 CI 0.08 to 15.41;  Analysis 3.7).

 
3.3.4 Specific symptoms

A series of specific symptoms related to mood, anxiety and tension were reported that occurred once medication had been instigated. It is feasible that these were part of the illness or that they may have been caused by the medication; whatever the cause, no apparent differences between pimozide and typical antipsychotic drugs were noted for depression in the short term (1 RCT n = 44, RR 0.61 CI 0.30 to 1.21), nor for improvement in psychomotor activity (1 RCT n = 20, RR 0.40 CI 0.10 to 1.60) or the presence of first-rank symptoms (1 RCT n = 44, RR 0.56 CI 0.24 to 1.29). By medium term, no differences between groups were apparent for anxiety/tension (3 RCTs n = 161, RR 0.67 CI 0.20 to 2.29) or for the presence of first-rank symptoms (1 RCT n = 44, RR 0.53 CI 0.25 to 1.11). However, the incidence of depression was greater with people receiving fluphenazine decanoate or thioridazine compared with pimozide (2 RCTs n = 74, RR 0.48 CI 0.26 to 0.89;  Analysis 3.8). This finding was statistically significant.

 
3.3.5 Use of additional medication

Results were equivocal at short term when pimozide was compared with chlorpromazine, flupenthixol or haloperidol (3 RCTs n = 114, RR 1.07 CI 0.77 to 1.49;  Analysis 3.9), and other antipsychotics were slightly favoured.

 

3.4 Adverse effects

 
3.4.1 Extrapyramidal adverse effects specific

Many effects are reported;see Figure 6 ( Analysis 3.10). Again, few clear differences are evident, and no results demonstrate statistical significance. For some outcomes, however, pimozide was more likely to cause adverse effects such as tremor by short term (6 RCTs n = 186, RR 1.52 CI 0.97 to 2.38) and by medium term (4 RCTs n = 175, RR 1.46 CI 0.68 to 3.11) when compared with both typical and atypical antipsychotics; these findings just failed to reach statistical significance (P = 0.07). This was also evident in results for tardive dyskinesia by medium term (4 RCTs n = 108, RR 1.27 CI 0.73 to 2.23) when compared with fluphenazine decanoate or trifluoperazine, and in results for akathisia by short term (4 RCTs n = 148, RR 1.18 CI 0.79 to 1.76).

 FigureFigure 6. Forest plot of comparison: 3 PIMOZIDE versus ANY ANTIPSYCHOTIC, outcome: 3.10 Adverse effects: 1. Extrapyramidal adverse effects - specific.

 
3.4.1.1 Akathisia

Pimozide appeared more likely to cause akathisia in the short term, but differences between groups did not reach statistical significance (4 RCTs n = 148, RR 1.18 CI 0.79 to 1.76). Medium-term results showed no real differences (5 RCTs n = 219, RR 1.17 CI 0.26 to 5.22) but demonstrated considerable heterogeneity (Chi² = 10.92, I² = 63%).

 
3.4.1.2 Acute dyskinesia

Pimozide was no more likely to cause acute dyskinesia than other antipsychotic drugs (haloperidol) in the short term (1 RCT n = 29, RR 2.68 CI 0.62 to 11.64).

 
3.4.1.3 Dystonia

Short-term results for dystonia appeared to favour those receiving other antipsychotic drugs, but the difference was not significant (2 RCTs n = 42, RR 0.78 CI 0.25 to 2.48). No difference was found between groups by medium term (2 RCTs n = 50, RR 1.30 CI 0.28 to 6.00).

 
3.4.1.4 Gait disturbance

Limited data show no real differences between groups for short-term (2 RCTs n = 59, RR 1.18 CI 0.16 to 8.62) or medium-term (1 RCT n = 38, RR 0.25 CI 0.03 to 2.04) follow-up.

 
3.4.1.5 Hypersalivation

Again, no clear differences between groups were found by short term (3 RCTs n = 115, RR 0.98 CI 0.31 to 3.12) or by medium term (3 RCTs n = 102, RR 0.88 CI 0.13 to 6.15); these results were slightly heterogeneic (Chi² = 3.11, I² = 36%).

 
3.4.1.6 Involuntary movements

One study found that pimozide was no more likely than other typical antipsychotic drugs to cause involuntary movements (1 RCT n = 87, RR 0.34 CI 0.01 to 8.14).

 
3.4.1.7 Parkinsonism

Results were equivocal and demonstrated that pimozide was no more likely than other antipsychotics to cause Parkinsonism by short term (1 RCT n = 56, RR 1.25 CI 0.37 to 4.17) and by medium term (2 RCTs n = 131, RR 2.12 CI 0.28 to 15.90).

 
3.4.1.8 Restlessness

Equivocal results were found in the short term (3 RCTs n = 101, RR 1.20 CI 0.62 to 2.34); medium-term results showed that more people in the pimozide group suffered from this side effect, but the difference failed to reach significance (3 RCTs n = 134, RR 2.59 CI 0.41 to 16.49).

 
3.4.1.9 Rigidity

Short-term results indicate that people who received pimozide were slightly more likely to experience this side effect (6 RCTs n = 186, RR 1.21 CI 0.71 to 2.05); these results display slight heterogeneity (Chi² = 7.51, I² = 33%). Medium-term data, however, are equivocal with only slight favour of other antipsychotics (5 RCTs n = 219, RR 1.12 CI 0.24 to 5.25); these results display considerable heterogeneity (Chi² = 8.31, I² = 52%).

 
3.4.1.10 Seizure

In the one study that reported data at short term, people receiving haloperidol were more likely to experience the side effect than people receiving pimozide (1 RCT n = 29, RR 0.36 CI 0.02 to 8.07).

 
3.4.1.11 Use of antiparkinsonian drugs

Limited data revealed no real differences in numbers of participants needing antIparkinson medication by the end of the studyin the short term (2 RCTs n = 67, RR 1.46 CI 0.54 to 3.94) or in the medium term (3 RCTs n = 116, RR 0.84 CI 0.41 to 1.72;  Analysis 3.10). Both sets of results demonstrated heterogeneity: only slightly at medium term (Chi² = 2.76, I² = 28%) but substantially at short term (Chi² = 2.88, I² = 65%).

 
3.4.2 Extrapyramidal adverse effects: average score (EPS), skewed

Skewed data from the EPS were presented from one study, indicating that people receiving pimozide experienced considerably more numerous extrapyramidal symptoms than people receiving chlorpromazine. However, these results must be interpreted with caution and are best considered by viewing  Analysis 3.11.

 
3.4.3 Death

Only one study reported death; this was described as a suicide in the control (chlorpromazine) group (1 RCT n = 51, RR 0.32 CI 0.01 to 7.53;  Analysis 3.12).

 
3.4.4 Anticholinergenic effects

Most of the data provided showed no differences between pimozide and other typical or atypical antipsychotics, including blurred vision (pimozide compared with fluphenazine decanoate, haloperidol, chlorpromazine, trifluoperazine or thioridazine), in the short term (4 RCTs n = 110, RR 0.98 CI 0.46 to 2.09) and in the medium term (5 RCTs n = 219, RR 0.98 CI 0.45 to 2.12); data were equivocal for sweating (when pimozide was compared with carpipramine or haloperidol; 2 RCTs n = 85, RR 2.16 CI 0.29 to 15.96) and in urinary retention at short term (pimozide compared with fluphenazine decanoate, carpipramine, haloperidol or trifluoperazine; 3 RCTs n = 121, RR 1.26 CI 0.28 to 5.67) and at medium term (2 RCTs n = 125, RR 0.26 CI 0.03 to 2.25). Data for dry mouth were equivocal at short term when pimozide was compared with fluphenazine decanoate, haloperidol, carpipramine or chlorpromazine (5 RCTs n = 166, RR 0.84 CI 0.54 to 1.31), but by medium term, data significantly favoured (P = 0.01) pimozide when compared with fluphenazine decanoate, trifluoperazine or thioridazine (4 RCTs n = 175, RR 0.44 CI 0.23 to 0.85;  Analysis 3.13).

 
3.4.5 Cardiovascular effects

No differences in various cardiovascular symptoms such as rise or fall in blood pressure and dizziness were noted between pimozide and other drugs. Pimozide was no more likely to cause a fall in blood pressure than were other antipsychotic drugs both in the short term and in the medium term.

Similar results were found for rise in blood pressure with equivocal data; pimozide also had no obvious effect on postural blood pressure. Short-term data for dizziness were similar for both groups, and again, no differences were found in the medium term.

Various cardiovascular adverse effects were recorded, each of which demonstrated equivocal results when pimozide was compared with other antipsychotics.

 
3.4.5.1 Chest pain

By medium term, when pimozide was comparedwith thioridazine, pimozide was slightly favoured (1 RCT n = 30, RR 0.33 CI 0.01 to 7.58).

 
3.4.5.2 Dizziness

Only slight favour of pimozide was seen by short term when compared with fluphenazine, chlorpromazine, haloperidol or carpipramine (5 RCTs n = 161, RR 0.73 CI 0.26 to 2.02). Again, by medium term, no difference between groups was seen when pimozide was compared with trifluoperazine or thioridazine (3 RCTs n = 161, RR 1.08 CI 0.34 to 3.43).

 
3.4.5.3 ECG changes:

A surprising lack of usable ECG data was reported. Kudo 1972 found no short-term differences in ECG changes (1 RCT n = 56, RR 0.33 CI 0.04 to 3.01).

 
3.4.5.4 Hypotension

By short term, results were equivocal (3 RCTs n = 114, RR 1.13, CI 0.50 to 2.59), with only slight favour for pimozide over fluphenazine at medium term in the one study that reported data (1 RCT n = 43, RR 0.32 CI 0.07 to 1.46). Postural hypotension was reported in two studies, with results demonstrating greater risk in the antipsychotics group (fluphenazine or haloperidol; 2 RCTs n = 42, RR 0.33 CI 0.06 to 1.92).

 
3.4.5.5 Hypertension

This was scarcely reported, and no difference between groups was found in the short term when pimozide was compared with carpipramine (1 RCT n = 56, RR 1.00 CI 0.07 to 15.21) nor in the medium term when it was compared with trifluoperazine (1 RCT n = 87, RR 0.51 CI 0.05 to 5.44).

 
3.4.5.6 Palpitations

Heart palpitations were reported in two studies that compared pimozide with haloperidol and or carpipramine; however, results are equivocal by short term (2 RCTs n = 78, RR 1.28 CI 0.31 to 5.28).

 
3.4.5.7 Substernal pain

This was reported in one study that compared pimozide with thioridazine, with slight favouring of pimozide by medium term (1 RCT n = 30, RR 0.33 CI 0.01 to 7.58;  Analysis 3.14).

 
3.4.6 Abnormal laboratory tests

Meta-analysis was not possible in this outcome, as data were reported only from single studies. No differences were seen between pimozide and control groups.

 
3.4.6.1 Haematological

Granulocytopenia was slightly more evident amongst people receiving fluphenazine than pimozide (1 RCT n = 20, RR 0.33 CI 0.02 to 7.32). No difference between groups with leukocytosis was reported at short term (1 RCT n = 20, RR 1.00 CI 0.34 to 2.93) and at medium term (1 RCT n = 51, RR 0.96 CI 0.06 to 14.55) when pimozide was compared with chlorpromazine or haloperidol. Slightly more people receiving pimozide experienced macrocytosis than those receiving haloperidol at short term (1 RCT n = 20, RR 5.00 CI 0.27 to 92.62); however, this was not a significant finding. By medium term, no difference could be seen in any abnormal data (1 RCT n = 30, RR 1.00 CI 0.31 to 3.28).

 
3.4.6.2 Liver function

No difference in any abnormal laboratory data was evident (1 RCT n = 30, RR 0.20 CI 0.03 to 1.51).

 
3.4.6.3 Renal function

No difference was noted between groups in results for albumin in urinalysis (1 RCT n = 20, RR 3.00 CI 0.14 to 65.90) and for granular cysts on urinalysis (1 RCT n = 20, RR 3.00 CI 0.14 to 65.90;  Analysis 3.15).

 
3.4.7 Central nervous system effects
 
3.4.7.1 Arousal

Results appeared to favour pimozide for both sedation and drowsiness; in the short term, pimozide appeared to present a lower incidence of sedation (1 RCT n = 42, RR 0.79 CI 0.47 to 1.34), with significant favour for pimozide evident at medium term (3 RCTs n = 155, RR 0.30 CI 0.12 to 0.72); these three studies consistently found pimozide less likely to cause sedation. At short term, pimozide was less likely to cause drowsiness when compared with haloperidol, chlorpromazine, fluphenazine or carpipramine (7 RCTs n = 226, RR 0.77 CI 0.60 to 1.00); these results demonstrated statistical significance (P = 0.05). Pimozide was also slightly favoured at medium term (5 RCTs n = 229, RR 0.61 CI 0.33 to 1.11); however, this finding was not significant. Insomnia was similar across groups in the short term (3 RCTs n = 101, RR 1.02 CI 0.67 to 1.55) and in the medium term (4 RCTs n = 181, RR 0.92 CI 0.35 to 2.39), with no differences evident.

Data for excitement were equivocal, with slight non-significant favouring of fluphenazine over pimozide in the short term (1 RCT n = 20, RR 3.00 CI 0.14 to 65.90) but slight favouring of pimozide over trifluoperazine in the medium term (1 RCT n = 87, RR 0.68 CI 0.12 to 3.88). Instances of faintness were greater amongst people receiving trifluoperazine than among those receiving pimozide in the one study that reported this adverse effect by medium term (1 RCT n = 20, RR 0.20 CI 0.01 to 3.70). No difference between groups was observed in the numbers of participants experiencing fatigue at short term (3 RCTs n = 101, RR 1.07 CI 0.57 to 2.00), nor levels of restlessness at short term (1 RCT n = 56, RR 1.20 CI 0.62 to 2.34) and at medium term (2 RCTs n = 95, RR 2.59 CI 0.41 to 16.49); however, medium-term data demonstrate considerable heterogeneity (Chi² = 2.50, I² = 60%;  Analysis 3.16).

 
3.4.7.2 Headaches

At short term, data indicate that slightly more people experienced headache as a side effect when given other antipsychotics (including haloperidol, chlorpromazine, carpipramine or fluphenazine decanoate (4 RCTs n = 161, RR 0.67 CI 0.21 to 2.07); these data are slightly heterogeneous (Chi² = 4.80, I² = 38%). Medium-term resultsseem to slightly favour the use of other antipsychotics (5 RCTs n = 219, RR 2.14 CI 0.84 to 5.45); these results are not statistically significant ( Analysis 3.16).

 
3.4.7.3 Memory defects

Clark 1975* presented equivocal data for memory defects (1 RCT n = 30, RR 0.33 CI 0.01 to 7.58;  Analysis 3.16).

 
3.4.8 Dermatological effects

Some studies reported specific adverse effects such as rash, dermatitis and itching, but no differences between pimozide and other drugs were clearly demonstrated for acne vulgaris (1 RCT n = 20, RR 3.00 CI 0.14 to 65.90), dermatitis (1 RCT n = 30, RR 1.00 CI 0.07 to 14.55), itching (2 RCTs n = 131, RR 0.61 CI 0.08 to 4.46) and rashes, short term (3 RCTs n = 145, RR 1.75 CI 0.47 to 6.53) or medium term (2 RCTs n = 125, RR 0.38 CI 0.11 to 1.31;  Analysis 3.17).

 
3.4.9 Endocrine effects

Endocrine effects were rarely recorded, and the small studies reporting amenorrhoea and galactorrhoea would have been very unlikely to find a difference between pimozide and other drugs, even if it exists. The limited data showed no differences between groups for short-term (1 RCT n = 56, RR 5.00 CI 0.25 to 99.67) and medium-term amenorrhoea (1 RCT n = 20, RR 3.00 CI 0.14 to 65.90), galactorrhoea (1 RCT n = 30, RR 0.33 CI 0.01 to 7.58) or acute cholecystitis (1 RCT n = 60, RR 2.81 CI 0.12 to 66.40;  Analysis 3.18).

 
3.4.10 Gastrointestinal effects

For a list of adverse effects including appetite change, nausea and weight change, pimozide was not clearly different from antipsychotic drugs ( Analysis 3.19).

 
3.4.10.1 Abdominal cramps

No difference was noted at medium term (3 RCTs n = 151, RR 1.97 CI 0.35 to 10.95).

 
3.4.10.2 Anal incontinence

No difference was noted at short term (1 RCT n = 20, RR 3.00 CI 0.14 to 65.90).

 
3.4.10.3 Appetite decrease

No difference was noted at short term (2 RCTs n = 85, RR 1.34 CI 0.22 to 8.07) or at medium term (2 RCTs n = 107, RR 1.05 CI 0.16 to 6.84); however, heterogeneity was considerable in the short term (Chi² = 2.73, I² = 63%).

 
3.4.10.4 Appetite increase

No difference was seen at short term (2 RCTs n = 85, RR 0.65 CI 0.15 to 2.83).

 
3.4.10.5 Constipation

A slighter higher occurrence of constipation was evident amongst people receiving other antipsychotics than pimozide by short term (5 RCTs n = 190, RR 0.64 CI 0.31 to 1.30) and by medium term (4 RCTs n = 132, RR 0.75 CI 0.30 to 1.88); however, these findings were not significant.

 
3.4.10.6 Diarrhoea

No difference was evident at short term (2 RCTs n = 76, RR 1.02 CI 0.11 to 9.33) or at medium term (1 RCT n = 30, RR 3.00 CI 0.13 to 68.26).

 
3.4.10.7 Nausea and vomiting

These were slightly more prevalent amongst people receiving other antipsychotics than amongst those given pimozide at short term (2 RCTs n = 76, RR 0.28 CI 0.05 to 1.61) and at medium term (5 RCTs n = 203, RR 0.64 CI 0.24 to 1.69).

 
3.4.11 Genitourinary effects

Each of three small trials reported a single symptom. Pimozide was no different from other drugs for the outcomes of anorgasmia (pimozide compared with thioridazine; 1 RCT n = 30, RR 3.00 CI 0.13 to 68.26), nocturia (pimozide compared with trifluoperazine; 1 RCT n = 87, RR 0.34 CI 0.01 to 8.14) and dysuria (pimozide compared with carpipramine; 1 RCT n = 56, RR 0.33 CI 0.01 to 7.85;  Analysis 3.20).

 
3.4.12 Other

When levels of intensified symptoms were compared, pimozide was slightly favoured in the short term when compared with carpipramine or trifluoperazine (2 RCTs n = 116, RR 0.48 CI 0.09 to 2.50) and in the medium term when compared with chlorpromazine or trifluoperazine (2 RCTs n = 92, RR 0.59 CI 0.27 to 1.28). In the meta-analysis, data were not significant but demonstrated slight favour of pimozide over the different antipsychotics (4 RCTs n = 208, RR 0.56 CI 0.28 to 1.14). With weight loss, no difference between groups was observed at short term (1 RCT n = 20, RR 0.33 CI 0.04 to 2.69) or at medium term (3 RCTs n = 161, RR 1.27 CI 0.42 to 3.82), nor was any difference reported between groups with weight gain at short term (1 RCT n = 43, RR 1.58 CI 0.16 to 16.17) or at medium term (4 RCTs n = 175, RR 1.46 CI 0.51 to 4.20;  Analysis 3.21).

 

3.5 Social functioning

 
3.5.1 Social functioning: average score (SBAS), skewed

Data from one study were skewed and are best examined through inspection ( Analysis 3.22). Results imply no difference between groups regarding levels of social functioning; however, these results must be interpreted with caution.

 

3.6 Leaving the study early

 
3.6.1 Because of relapse

Relapse during the study was another reason for early attrition in some trials, but short-term data from one trial (Chouinard 1970) showed no difference between pimozide and trifluoperazine at three months (1 RCT n = 20, RR 3.00 CI 0.14 to 65.90). Three trials presented medium-term data, and results appear to slightly favour pimozide (3 RCTs n = 76, RR 0.88 CI 0.28 to 2.78;  Analysis 3.23).

 
3.6.2 Because of adverse effects

When adverse effects were cited as the cause of attrition, data from short-term trials (6 RCTs n = 234, RR 0.80 CI 0.32 to 2.01) and from medium-term trials (7 RCTs n = 252, RR 0.99 CI 0.57 to 1.72;  Analysis 3.24) were equivocal. The latter data were derived from two trials in which pimozide was compared with depot fluphenazine.

 
3.6.3 For any reason

Eleven studies found no difference in attrition between pimozide and antipsychotic drugs in the short term when the reason for attrition was not made explicit (11 RCTs n = 368, RR 1.23 CI 0.63 to 2.40). In the medium term, fourteen trials still found no clear difference (14 RCTs n = 523, RR 0.87 CI 0.69 to 1.09), even when the five studies comparing pimozide with depot fluphenazine decanoate were analysed separately. All data are homogeneous. Finally, Abuzzahab 1980*, in a 15-month study, found no long-term differences between pimozide and oral fluphenazine (1 RCT n = 62, RR 1.00 CI 0.60 to 1.67;  Analysis 3.25).

 

4. Comparison 4: PIMOZIDE + ANY ANTIPSYCHOTIC versus ANY ANTIPSYCHOTIC

Results for this comparison came from three studies (n = 142), with most of the data reported from two studies (De Ronchi 1996; Friedman 2011), which reported results for both primary outcomes of mental state and extrapyramidal adverse effects. The third study (Nishikawa 1985) provided data on global state relapse and leaving the study early. Only data on leaving the study early were pooled, as each study reported all other outcomes separately. Most reported data are skewed and should be interpreted with caution.

 

4.1 Global state

 
4.1.1 Relapse: clinical diagnosis

In the one small study that reported this outcome, a significantly greater number of participants receiving thioridazine alone experienced relapse (as clinically diagnosed in the study) compared with people receiving pimozide combined with the same antipsychotic at medium term (1 RCT n = 69, RR 0.28 CI 0.15 to 0.50;  Analysis 4.1); this was a statistically significant finding (P = 0.00001;  Analysis 4.1).

 

4.2 Leaving the study early

 
4.2.1 Any reason

Levels of attrition for any reason seemed to favour the pimozide+antipsychotic treatment group over the antipsychotics group at medium term (1 RCT n = 69, RR 0.56 CI 0.42 to 0.74;  Analysis 4.2).

 

5. Comparison 5. PIMOZIDE + ANY ANTIPSYCHOTIC versus ANY ANTIPSYCHOTIC + PLACEBO

 

5.1 Global state

 
5.1.1 Average score (CGI), skewed

Data were skewed for average change score when the CGI was used in one study ( Analysis 5.1).

 
5.1.2 Average score (CGI-S)

Data show no difference in CGI-S scores at short term (n = 28, RR 0.00 CI -0.49 to 0.49;  Analysis 5.2).

 
5.1.3 Average score (CGI-I)

Data from a single study show significance suggesting greater improvement just with people receiving clozapine+placebo (P = 0.04) at short term (n = 28, RR 0.60 CI 0.01 to 1.19;  Analysis 5.3).

 

5.2 Mental state

 
5.2.1 Average score (BPRS)

Data from the same small single study again show slight significance, suggesting greater improvement amongst people receiving clozapine+placebo (P = 0.03) at short term (n = 28, RR 1.70 CI 0.14 to 3.26;  Analysis 5.4).

 
5.2.2 Average score (BPRS psychosis subscale)

Much greater statistically significant results were found in the same small study, again implying greater improvement with the clozapine+placebo combination (n = 28, RR 1.70 CI 0.96 to 2..44;  Analysis 5.5).

 
5.2.3 Average score (PANSS), skewed

Data from a single study were skewed for this outcome and are best considered by inspection ( Analysis 5.6).

 
5.2.4 Average score (SANS)

Results were equivocal between groups at short term when SANS was used (n = 28, RR 1.00 CI -3.37 to 5.37;  Analysis 5.7).

 

5.3 Adverse effects

 
5.3.1 Extrapyramidal adverse effects: specific

Results from a single study seem to suggest that more people receiving combined pimozide/clozapine experienced hypersalivation than people receiving combined clozapine/placebo at medium term (1 RCT n = 53, RR 2.99 CI 0.89 to 10.04;  Analysis 5.8); however, this finding was not statistically significant.

 
5.3.2 Extrapyramidal adverse effects: average score (ESRS), skewed

EPS symptoms were measured in a single study using the ESRS; results are skewed and, again, show no meaningful differences between groups ( Analysis 5.9).

 
5.3.3 Cardiovascular effects

Results were equivocal at medium term with occurrences of bigeminy (1 RCT n = 53, RR 0.37 CI 0.02 to 8.73) or hypotension (1 RCT n = 53, RR 0.37 CI 0.02 to 8.73) only slightly more prevalent amongst people given clozapine ( Analysis 5.10).

 
5.3.4 Cardiovascular effects: average score

Data from a small study show that QTc levels were far greater in people receiving combined pimozide+clozapine than in people receiving clozapine+placebo (n = 28, RR 7.40 CI 3.62 to 11.17;  Analysis 5.11).

 
5.3.5 Cardiovascular effects: average score, skewed

All data for changes in cardiovascular activity, including diastolic blood pressure, heart rate (bpm), QTc and systolic blood pressure, are skewed and are best inspected by using an additional table ( Analysis 5.12).

 
5.3.6 Abnormal laboratory tests: average score, skewed

Data for abnormal laboratory tests, including changes in plasma glucose level and cholesterol level, are skewed and are presented in a separate table ( Analysis 5.13).

 
5.3.7 Other specific effects

One study reported a higher level of 'intensified symptoms' at medium term when participants received combined clozapine/placebo rather than combined pimozide/clozapine (1 RCT n = 53, RR 0.22 CI 0.01 to 4.44); however, this finding was not significant ( Analysis 5.14).

 

5.4 Quality of life

 
5.4.1 Average score (SLOF), skewed

All quality of life data are skewed and are best presented in an additional table ( Analysis 5.15). Results indicate a greater level of functioning in interpersonal relationships, social acceptability and work skills at medium term when combined pimozide/clozapine is compared with clozapine/placebo. Again, these results are skewed and must be interpreted with caution.

 

5.5 Leaving the study early

 
5.5.1 Because of adverse effects

Results are equivocal for numbers of participants leaving the study early because of adverse effects at short term (1 RCT n = 28, RR 3.00 CI 0.13 to 67.91) and at medium term (1 RCT n = 53, RR 0.56 CI 0.11 to 2.80); results are equivocal overall (2 RCTs n = 81, RR 0.85 CI 0.22 to 3.25;  Analysis 5.16).

 
5.5.2 For any reason

Levels of attrition for any reason were again equivocal at short term and at medium term (2 RCTs n = 81, RR 1.09 CI 0.35 to 3.33;  Analysis 5.17).

 

6. Comparison 6. PIMOZIDE + ANY ANTIPSYCHOTIC versus ANY ANTIPSYCHOTIC + ANY ANTIPSYCHOTIC

 

6.1 Global state

 
6.1.1 No improvement

In another small study, lower levels of improvement were seen amongst people receiving combined pimozide/haloperidol compared with people receiving combined haloperidol/levosulpiride at short term (1 RCT n = 20, RR 5.00 CI 0.70 to 35.50); however, this finding was not significant ( Analysis 6.1).

 
6.1.2 Average score (CGI-S)

Data reported from another single study suggest greater global improvement amongst people receiving haloperidol/levosulpiride than amongst those receiving combined pimozide/haloperidol at short term using the CGI-Severity of illness (1 RCT n = 20, MD 2.50 CI 1.30 to 3.70;  Analysis 6.2).

 
6.1.3 Average score (CGI-I)

The same study reported global improvement on the CGI-I amongst people receiving haloperidol/levosulpiride (1 RCT n = 20, MD 1.50 CI 0.93 to 2.07; Analysis 4.5) at short term. These results, although demonstrating statistical significance (P = 0.0001 or greater), should be interpreted with caution because of the small sample size ( Analysis 6.3).

 

6.2 Mental state

 
6.2.1 Average score (SANS)

Short-term data from the same study using the SANS, again, indicate a higher level of improvement in negative symptoms when people received combined haloperidol/levosulpiride rather than combined pimozide/haloperidol (1 RCT n = 20, MD 24.00 CI 5.98 to 42.02). Results for SANS demonstrate statistical significance (P = 0.009); however, caution is needed when these results are interpreted because of their limited power ( Analysis 6.4).

 
6.2.2 Average score (SAPS), skewed

Data from a single study were skewed when SAPS was used; however they indicate a higher level of improvement in positive symptoms when people received combined haloperidol/levosulpiride rather than combined pimozide/haloperidol at short term (see  Analysis 6.5).

 

6.3 Adverse effects

 
6.3.1 Extrapyramidal adverse effects: average score, skewed

Furthermore, other EPS symptoms reported by another single study suggest no difference in levels of akathisia, akinesia, rigidity and tremor at short term; however, results are skewed and are best presented in an additional table for inspection ( Analysis 6.6).

 
6.3.2 Anticholinergenic effects: average score, skewed

A single study reported skewed data that may suggest greater instances of dry mouth and blurred vision amongst people receiving combined pimozide/haloperidol rather than combined levosulpiride/haloperidol at short term ( Analysis 6.7).

 
6.3.3 Cardiovascular effects: average endpoint score, skewed

Furthermore, instances of dizziness, hypotension and tachycardia present skewed data that are suitably presented in another separate additional table ( Analysis 6.8).

 
6.3.4 Autonomic effects: average score, skewed

Skewed data for autonomic effects are made available from a single study; upon inspection, no difference between groups is apparent ( Analysis 6.9).

 
6.3.5 Central nervous system effects: average score, skewed

Again, central nervous system effects from a single study are skewed to report the data; no difference is implied between groups in drowsiness, headache and insomnia at short term ( Analysis 6.10).

 
6.3.6 Endocrine effects: average score, skewed

Endocrine effects, including amenorrhoea and galactorrhoea, were reported in a single study; however, data are skewed and must be interpreted with caution ( Analysis 6.11).

 
6.3.7 Gastrointestinal effects: average score, skewed

With skewed data from a single study, no clear difference between groups was evident with effects such as constipation, nausea, weight loss and weight gain ( Analysis 6.12).

 
6.3.8 Gastrointestinal effects

In the same study non-skewed data demonstrated no difference by short term amongst participants who experienced constipation (1 RCT n = 20, RR 3.00 CI 0.14 to 65.90;  Analysis 6.13).

 
6.3.9 Other specific adverse effects

Again, one study reported mammary tension by short term, which occurred more frequently amongst people receiving combined pimozide/haloperidol than amongst people receiving combined levosulpiride/haloperidol (1 RCT n = 20, RR 5.00 CI 0.70 to 35.50;  Analysis 6.14); however this was not a significant finding.

 
6.3.10 Other specific effects: average score, skewed

In addition, skewed data were presented for gynaecomastia in a single study; results imply no clear difference between groups and are presented in a separate table ( Analysis 6.15).

 

6.4 Leaving the study early

 
6.4.1 Because of adverse effects

Results are equivocal for numbers of participants leaving the study early because of adverse effects at short term (1 RCT n = 20, RR 0.33 CI 0.02 to 7.32;  Analysis 6.16).

 
6.4.2 For any reason

Again, results are equivocal at short term (3 RCTs n = 20, RR 0.33 CI 0.02 to 7.32;  Analysis 6.17).

 

7. Comparison 7. SENSITIVITY ANALYSIS: PIMOZIDE versus ANY ANTIPSYCHOTIC

 

7.1 Implication of randomisation and risk of bias

For our primary outcomes, only three studies reported data; therefore, a meaningful sensitivity analysis for randomisation is weak. For mental state: relapse, no indication in the results suggest that a study rated at 'high risk' was likely to overestimate the effects of the intervention ( Analysis 7.1). This was also the case for our additional primary outcome of Parkinsonism ( Analysis 7.2). No difference was seen in the direction of estimates of effects when trials that were judged to be at high risk of bias across one or more of the domains of randomisation were removed from the meta-analysis (where relevant).

 

7.2 Assumptions for lost binary data

Sensitivity analysis in which those who left the study were not assumed to have a bad outcome did not substantially change any of the main outcomes.

 

7.3 Imputed values

No values included in the data and analysis were imputed.

 

7.4 Fixed-effect and random-effects

When primary outcome data were synthesised using a random-effects model, no significant differences were noted in the estimate of the effect.

 

8. Publication bias

Funnel plots were planned to investigate the possibility of publication bias (see Methods). The small study sizes led us to seek expert advice (see 'Acknowledgements'). Such plots are not powerful investigative tools and are further weakened when study size shows little variation (Egger 1997). Global state: No improvement data from the comparison of pimozide versus any antipsychotic were used to investigate whether systematic small trial bias was evident in a funnel plot analysis (Figure 2). Although the plot appears symmetrical (indicating an absence of bias), it should be remembered that a small number of studies was included, and it is difficult to draw any meaningful conclusions from the data. We think it inadvisable to read too much into this exploratory, low-powered technique of investigation. More studies with a wide spread of findings are needed.

 

Discussion

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms
 

1. Comparison 1: PIMOZIDE versus PLACEBO

Please see  Summary of findings for the main comparison. Data are very limited, allowing for few confident conclusions. Pimozide does appear better than placebo for preventing relapse by medium term, with only one small study providing data. Acceptability of treatment, as measured indirectly by leaving the study early, appeared the same for placebo for the first three months but favoured pimozide in the longer term. All adverse effects data are subject to random error, making confidence in the outcomes impossible.

 

2. Comparison 2: PIMOZIDE versus PLACEBO: WITHDRAWAL STUDY

The one withdrawal study (Denijs 1973; n = 40) comparing continuation of pimozide with its double-blind substitution by placebo suggested that once a person is stabilised by taking pimozide, its withdrawal will usually result in the need for supplementary antipsychotic treatment. One death was reported with pimozide, but this death was reported by the authors as not related to treatment.

Data from this small study support the impression that pimozide is an effective antipsychotic. Fewer people continuing pimozide required additional medication when compared with those withdrawn from it.

 

3. Comparison 3: PIMOZIDE versus ANY ANTIPSYCHOTIC

Please see  Summary of findings 2.

 

3.1 Global state

Generally, pimozide had similar efficacy to other antipsychotic drugs in preventing relapse and various measures of mental state. The outcome of admission is reported in only one study, with oral pimozide not clearly different from depot fluphenazine decanoate. For the unusual group of people recruited to such a trial, the preparations and the drugs may be interchangeable and tailored to clients’ needs and preferences. This finding is unlikely to be generalisable to real-world mental healthcare.

All last observation carried forward analyses contain assumptions regarding the stability of the state of those who left the study early. It is difficult to know how these assumptions may have affected the final results, but analyses of various measures of global state consistently failed to reveal clear differences between the experimental drug and older typical antipsychotic drugs such as the 'benchmark' drug chlorpromazine (Adams 2007).

 

3.2 Mental state

Overall, data support the clinical efficacy of pimozide but not its superiority over more typical drugs.

 

3.3 Adverse effects

One control group death was due to suicide (Kolivakis 1974); this is consistent with a lifetime rate of about 11%, as would be expected for this group. The lifetime incidence of suicide amongst people suffering from schizophrenia is 10% to 13% (Caldwell 1992). The use of high doses of antipsychotic drugs has been associated with sudden death (Jusic 1994), but no sudden or cardiac deaths were reported in the included studies, possibly because. most participants did not receive doses above present recommended limits. Case reports of sudden death amongst those taking pimozide were associated with doses higher than 20 mg (CSM 1990). Alternatively, this meta-analysis may not have had sufficient power to detect what might be a rare event.

No difference was found between the overall tolerability of pimozide and that of other drugs as measured indirectly by leaving the study because of adverse events.

A great number of adverse effects were listed in the included studies, but few showed clear differences between pimozide and other drugs. No differences were reported for the rare outcome of death, the several types of abnormal laboratory tests and cardiovascular problems. The latter include ECG changes, but these were reported in only one small study. The caution that surrounds the use of pimozide for people with cardiac problems (BNF 2012) is not based on trial-derived data.

Pimozide was found more likely to cause Parkinsonian tremor (number needed to treat for an additional harmful outcome (NNTH) 8) than typical antipsychotic drugs in the first three months of use. The difference in extrapyramidal adverse events was not maintained in the longer term (three months to one year). These limited data suggest that pimozide has clinically important extrapyramidal adverse effects.

In the medium term, pimozide was less sedating than the typical drugs (NNTH 4 CI 4 to 7). These differences are consistent with pimozide having more specific dopamine blocking activity than the antipsychotic drugs with which it was compared.

 

3.4 Leaving the study early

This, the most data-replete result of the review, suggests no difference between pimozide and other drugs. Leaving the study early may be an indirect measure of acceptability of treatment. It has been suggested that because of its longer half-life, pimozide may be useful for people who find it difficult to take tablets on a regular basis (McCreadie 1987). Data from trials do not support this. On the other hand, in the medium term, pimozide did provide similar results when compared with fluphenazine depot in five studies (n = 189). In two of the comparisons with fluphenazine decanoate depot (n = 63), pimozide was given on an intermittent basis, thus taking advantage of its long half-life. Randomised trials that include relatively compliant people are very limited in their ability to investigate the value of depot medication for those who find compliance with medication difficult (Quraishi 1999), and few claims should be made for pimozide on the basis of the limited data presented in this review. Only 7% of more than 500 people left the study early in the short term, and 35% by up to a year. This supports the theory that people in these trials were able to stick to study protocol, and differences in the outcome of compliance would not have been expected.

 

4. Comparison 4. PIMOZIDE + ANY ANTIPSYCHOTIC versus ANY ANTIPSYCHOTIC

Please see  Summary of findings 3. Data were scarce, but in the one study that made this comparison, pimozide when used in combination with another antipsychotic showed significantly reduced relapse rates and earlier losses than were seen when the same antipsychotic was given alone. This was demonstrated in one small study (n = 69), and the result was statistically significant (P = 0.00001).

 

5. Comparison 5. PIMOZIDE + ANY ANTIPSYCHOTIC versus ANY ANTIPSYCHOTIC + PLACEBO

Please see  Summary of findings 4. The two small studies that presented data for this comparison used the same intervention (pimozide+clozapine vus clozapine+placebo); however, results were largely presented individually in data and by analyses, largely as the result of lack of uniform outcome measurements and considerable quantities of skewed data. Therefore, meta-analysis was an option only for the outcome of leaving the study early, which presented equivocal data. This comparison was made by the two most recent studies included in this review; larger studies will be needed to further clarify these results.

 

6. Comparison 6. PIMOZIDE + ANY ANTIPSYCHOTIC versus ANY ANTIPSYCHOTIC + ANY ANTIPSYCHOTIC

Please see  Summary of findings 5.

With only one small study providing data for all outcomes (De Ronchi 1996), it is difficult to make any sound judgement of the potential efficacy of augmented pimozide under this comparison, particularly because most reported data suggest no differences between groups or are skewed. All statistically significant data that are reported suggest greater improvement in global state and mental state scores when participants received levosulpiride+haloperidol, as opposed to the pimozide+haloperidol combination. Further research is needed if any confidence in these results is to be attained.

 

7. Overall conclusion

Although shortcomings and gaps in the data are evident, overall consistency over different outcomes and time scales is sufficient to confirm that pimozide is an antipsychotic of similar efficacy to other more commonly used antipsychotic drugs, such as chlorpromazine, for people with schizophrenia. It may have a slightly different side effect profile from typical antipsychotic drugs. Data related to delusional disorder and mono-delusional states are not trial-based.

 

Overall completeness and applicability of evidence

 

1. Completeness

This review does report how the trials considered outcomes such as service use, global scores, mental state and adverse effects. However, data on any one outcome are few and power is often limited. Most data are short to medium term, and we obtained no data on quality of life, satisfaction with care or clear functioning outcomes. No people were found to have a specific diagnosis of delusional disorder. This is disappointing in the light of claims made for the drug (Kaplan 1994; Opler 1995).

 

2. Applicability

Overall, generalisability was good. All 32 studies included people with schizophrenia who would be recognisable in everyday practice. In most of the studies, the diagnosis was clinical, but a few used rigorous criteria. Participants with acute illness and those with chronic illness participated, and studies were undertaken in both hospital and community settings. The daily doses of pimozide largely reflect present practice, although five studies did employ higher levels than would be expected nowadays. Most outcomes reported are, we believe, accessible to clinicians and to recipients of care.

Most of the included studies were published between 1970 and 1987, with the exception of one study (Friedman 2011), which may reduce applicability to current practice. People with both schizophrenia and substance misuse were frequently excluded; this may reduce the applicability of findings because co-existence of the two problems is common (Turner 1990). Non-Western cultures were represented by only two small studies; therefore, applicability to those in the developing world may be limited.

 

Quality of the evidence

See also  Summary of findings for the main comparison;  Summary of findings 2 and  Summary of findings 3.

This review currently includes 32 studies with 1277 participants with ages ranging from 16 to 80 years and average age of 43 years. The sexes appeared evenly distributed, with 693 males and 584 females. The included studies were published over a long period (from 1971 to 2012) and in different settings (e.g. inpatients and outpatients). Many trials were completed within four months, and only one study lasted three years. The shortest study was conducted for four weeks, and the longest study had a duration of three years. Most participants had received a diagnosis of schizophrenia. However, the diagnostic criteria were not specified in 58% of the included studies, in which diagnosis was made pragmatically. Other studies used Diagnostic and Statistical Manual of Mental Disorders (DSM) III, DSM IV, International Classification of Diseases (ICD) and Research Diagnostic Criteria (RDC) as the criteria for diagnosis. The dose of pimozide ranged from 1 to 70 mg per day, with an average dose of 45.7 mg. Most of the studies were reported to be randomised and to use double-blind technique. However, details were not provided. Only four of the studies described randomisation adequately. Outcomes, participants leaving the study early and statistical data were poorly reported in most studies.

Important methodological limitations were present in the trials included in this review. Some were carried out before the CONSORT (Consolidated Standards of Reporting Trials) statement of 2001 was issued (Moher 2001), thus the level of reporting currently expected was not available in the papers describing these studies, particularly with regard to allocation concealment, randomisation and blinding processes.

Almost all studies were reported as randomised and double-blind, but for most of them, details were not presented. Therefore it is unclear whether the studies were adequately randomised, whether treatment allocation was concealed and whether blinding worked. For each study in the review, areas of unclear bias were identified and are listed in the risk of bias tables under each table of characteristics of included studies; they are also graphically summarised in Figure 4 and Figure 5. Data on participants leaving the study early were not reported in many cases, and statistical reporting was incomplete in almost all studies.

Many studies received funding from various sources, including pharmaceutical companies. In conclusion, the quality of evidence remains poor. 

 

Potential biases in the review process

The review authors sought to adhere to the protocol by independently inspecting citations and full articles of potentially relevant studies. Furthermore, the review authors independently extracted data onto standard simple forms; however, it was required that they meet for discussion when inconsistencies or disagreements arose regarding the data presented. We contacted study authors when data were incoherent or were difficult to interpret.

 

Agreements and disagreements with other studies or reviews

We are not aware of other systematic reviews conducted to evaluate the effects of pimozide on schizophrenia or related psychosis. This review updates, improves and largely concurs with the past version of the same review (Rathbone 2007).

 

Authors' conclusions

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms

 

Implications for practice
1. For people with schizophrenia

Pimozide is probably as effective as other commonly used antipsychotic treatments for schizophrenia. It may therefore be a matter of personal preference as to which treatment is best to select. For those with delusional disorder, no strong evidence suggests that it is an effective treatment or that it is preferable to other antipsychotic drugs, so those with delusional disorder might ask whether other treatments are better supported by evidence. The review authors found pimozide to be a safe drug as long as the currently recommended levels (20 mg/d) are not exceeded. ECG monitoring is needed because use of pimozide has been associated with sudden death at doses higher than are now used. People who are not happy with this situation might prefer an alternative. Pimozide may be less prone to cause sleepiness than typical antipsychotic drugs, but this observation has to be weighed against its possible propensity to cause increased tremor, which may need to be relieved by additional antiparkinsonian medication.

2. For clinicians

Pimozide appears to be a treatment of similar efficacy to typical antipsychotic drugs for those with schizophrenia. However, the often made claim that pimozide is a particularly useful treatment for delusional disorder is not based on trial evidence, and no evidence of its efficacy has been found amongst those whose illness is dominated by negative symptoms. Oral pimozide might be useful for those who find depot medication intolerable. Pimozide may be more likely than typical drugs to cause parkinsonian tremor and the need for antiparkinsonian medication, but it appears to be less sedating than other drugs. Overall tolerability for patients (measured indirectly) appears the same as for typical antipsychotic drugs, so clinicians should consider individual susceptibility to adverse effects and patient preference. Sudden or cardiac death was not reported. Pimozide appears to be safe in doses below 20 mg/d. Unfortunately, very few data have been obtained from controlled trials on ECG changes to guide clinicians on optimal monitoring. Evidence is insufficient to allow the review authors to determine the frequency of asymptomatic ECG changes that might require the drug to be stopped in clinical practice. Furthermore, evidence is insufficient to suggest that pimozide augmentation with any antipsychotic may lead to better results in reducing relapse rates and improving mental state or overall quality of life.

3. For managers/policy makers

Pimozide is inexpensive and appears as effective as other 'first-line', typical antipsychotic drugs. It seems to have a different adverse effect profile when compared with these other drugs and therefore may be preferred by certain people. However, its use does require ECG monitoring, which could be considered an extra cost in terms of staff time.

 
Implications for research
1. General

The trials reviewed predated the CONSORT statement (Begg 1996; Moher 2001). Had this statement been anticipated, a greater quantity of data would have been available to inform practice. Allocation concealment gives the assurance that selection bias is kept to the minimum and should be properly executed and described. Well-reported and well-tested blinding could have encouraged confidence in the control of performance and detection bias. It is important to know how many, and from which groups, people were withdrawn if exclusion bias is to be evaluated. Raters should be independent of treatment. It would have been helpful if authors had presented data in a useful manner that reflects the association between intervention and outcome, for example, relative risk, odds ratio, risk or mean differences, as well as raw numbers. Binary outcomes should be calculated in preference to continuous results, as they are easier to interpret. If P values are used, the exact values should be reported. Trials should report service utilisation data, as well as information on satisfaction with care and economic outcomes (see  Table 1).

2. Specific

The efficacy of pimozide seems comparable with that of other typical antipsychotic drugs, but it has not been tested adequately against placebo, probably because licensing requirements were less stringent at the time pimozide was developed. For example, it has never been tested in a randomised trial for delusional disorder in spite of the claim that it is particularly useful for this condition. This may be a result of the relative rarity of this condition (Winokur 1977), or it may be associated with difficulty in obtaining co-operation from this group of people, but claims should be supported by high-grade evidence. The claim that pimozide is useful for people with negative symptoms has not been tested in a trial in which these were properly measured with a well-validated scale. Also, good quality ECG data from clinical trials are insufficient to inform clinicians on the optimal monitoring strategy.

As with so many antipsychotic drugs, the efficacy of this widely available drug has never really been adequately tested in real-world conditions. If energy and funding can be found, good, clinically meaningful trials are justified, probably specifically for those with monosymptomatic hypochondriacal psychosis. This would be a most difficult study to undertake and would necessitate wide collaboration on outcomes of agreed relevance. Nevertheless, this group of people can be most disabled, and should pimozide have a role here, its use would indeed be most valuable. A suggested design is given in  Table 1.

 

Acknowledgements

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms

CEA - Clive Adams (Cochrane Schizophrenia Group, UK) undertook searches and gave invaluable help in preparing the protocol, selecting and acquiring studies,helping with extracting and summating data and producing the report. His enthusiasm and energy are infectious! We wish to thank Dr. Jayaram Mahesh for his support and feedback throughout the most recent updating process.

Jo Wood (Janssen-Cilag, UK) was most helpful in searching the company's database and supplying articles. Mattias Egger (University of Bristol, UK) gave advice on the validity of funnel plots and allowed perusal of a yet unpublished study on this matter, for which we are grateful.

Several letters were sent by the review authors to study authors, asking for extra information related to their trials. RG McCreadie (Crighton Royal Hospital, Dumfries, UK) was very kind in responding and offering access to unpublished data. We thank Thomas Barnes, who provided further information related to his study included in this review, as did Stephen Platt, who provided us with details of the SBAS rating scale used in the same study. We would also like to thank Diana De Ronchi for her correspondence and assistance.

Karen Perkins and other staff of the well-organised medical library (St. Andrew's Hospital, Northampton, UK) gave invaluable help in obtaining articles and using computer databases. Also thanks to Di Wyman (St. Andrew's Hospital, Northampton, UK) for her patient and efficient administrative help. Thanks are also due to Dr. Nevison Andrews (St. Andrew's Hospital, Northampton, UK) for support and encouragement, and to Dr. Staley (St. Andrew's Hospital, Northampton, UK), likewise for support and encouragement to pursue research projects.

We are really sorry not to include Alec Sultana and Tom McMonagle as co-authors. We would happily do so but have lost all contact with these authors. Elaine Sultana (spouse) helped with correspondence and was always tolerant and supportive during the hours of computer work. Alec Sultana prepared the first protocol, selected studies for the first version of this review, extracted data and helped write the report. Tom McMonagle also provided support during writing of protocol, study selection, data extraction, summation of data and report writing. John Rathbone (update 2005) selected studies, extracted data and assisted with report writing.

 

Data and analyses

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms
Download statistical data

 
Comparison 1. PIMOZIDE versus PLACEBO

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Global state: 1. Relapse - clinical diagnoses1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    1.1 medium term (3-12 months)
120Risk Ratio (M-H, Fixed, 95% CI)0.22 [0.03, 1.78]

 2 Global state: 2. No improvement3102Risk Ratio (M-H, Random, 95% CI)0.84 [0.69, 1.02]

    2.1 short term (<3 months)
240Risk Ratio (M-H, Random, 95% CI)0.61 [0.15, 2.45]

    2.2 medium term (3-12 months)
262Risk Ratio (M-H, Random, 95% CI)0.84 [0.67, 1.06]

 3 Mental state: 1. Specific symptoms2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    3.1 anxiety/tension - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)2.06 [0.09, 46.11]

    3.2 auditory hallucinations - short term (<3 months)
119Risk Ratio (M-H, Fixed, 95% CI)3.3 [0.15, 72.08]

    3.3 depression - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.67 [0.05, 9.47]

 4 Adverse effects: 1. Extrapyramidal adverse effects2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    4.1 dystonia - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.67 [0.05, 9.47]

    4.2 akathisia - short term (<3 months)
119Risk Ratio (M-H, Fixed, 95% CI)5.5 [0.30, 101.28]

    4.3 akathisia - medium term (<3 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.22 [0.03, 1.85]

    4.4 restlessness - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)2.0 [0.24, 16.61]

    4.5 rigidity - short term (<3 months)
119Risk Ratio (M-H, Fixed, 95% CI)5.5 [0.30, 101.28]

    4.6 rigidity - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)1.33 [0.14, 12.82]

    4.7 tremor - short term (<3 months)
119Risk Ratio (M-H, Fixed, 95% CI)3.3 [0.15, 72.08]

    4.8 tremor - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.20, 4.95]

 5 Adverse effects: 2. Anticholinergic effects1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    5.1 blurred vision - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.20, 4.95]

    5.2 dry mouth - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.23 [0.01, 5.12]

 6 Adverse effects: 3. Cardiovascular effects2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    6.1 blood pressure increase - short term (<3 months)
119Risk Ratio (M-H, Fixed, 95% CI)3.3 [0.15, 72.08]

    6.2 chest pain - medium term (<3 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    6.3 dizziness - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.67 [0.11, 3.99]

    6.4 ECG: primary changes in T waves - short term (<3 months)
119Risk Ratio (M-H, Fixed, 95% CI)3.3 [0.15, 72.08]

 7 Adverse effects: 4. Abnormal laboratory tests2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    7.1 haematological - any abnormal data - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.53 [0.19, 1.51]

    7.2 haematological - increase white cell count >12,000 - short term (<3 months)
119Risk Ratio (M-H, Fixed, 95% CI)0.56 [0.06, 5.14]

    7.3 liver function - any abnormal data - less than 12 months
125Risk Ratio (M-H, Fixed, 95% CI)2.06 [0.09, 46.11]

    7.4 liver function - increase alkaline phosphate > 35 I.U. - short term (<3 months)
119Risk Ratio (M-H, Fixed, 95% CI)1.67 [0.36, 7.82]

    7.5 renal - abnormal blood urea nitrogen balance - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.23 [0.01, 5.12]

 8 Adverse effects: 5. Central nervous system effects2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    8.1 drowsiness - short term (<3 months)
119Risk Ratio (M-H, Fixed, 95% CI)0.22 [0.01, 4.05]

    8.2 sedation - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.03, 3.20]

    8.3 headache - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)2.33 [0.60, 9.02]

    8.4 insomnia - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.58 [0.31, 1.09]

    8.5 memory defects - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 9 Adverse effects: 6. Dermatological effects1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    9.1 dermatitis - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.03, 3.20]

 10 Adverse effects: 7. Endocrine effects1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    10.1 galactorrhoea - medium term (3-12 months)
130Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 11 Adverse effects: 8. Gastrointestinal effects2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    11.1 acute cholecystitis - short term (<3 months)
145Risk Ratio (M-H, Fixed, 95% CI)1.41 [0.06, 32.53]

    11.2 constipation - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    11.3 diarrhoea - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.22 [0.03, 1.85]

    11.4 nausea - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.22 [0.03, 1.85]

 12 Adverse effects: 9. Other4262Risk Ratio (M-H, Fixed, 95% CI)0.57 [0.34, 0.96]

    12.1 anorgasmia - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)2.06 [0.09, 46.11]

    12.2 intensified symptoms - short term (<3 months)
145Risk Ratio (M-H, Fixed, 95% CI)0.09 [0.00, 1.83]

    12.3 intensified symptoms - medium term (3-12 months)
141Risk Ratio (M-H, Fixed, 95% CI)0.42 [0.15, 1.16]

    12.4 weight loss - short term (<3 months)
119Risk Ratio (M-H, Fixed, 95% CI)0.63 [0.28, 1.46]

    12.5 weight loss - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.44 [0.09, 2.20]

    12.6 weight gain - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)3.44 [0.18, 64.88]

    12.7 ocular pigment deposit increase - short term (<3 months)
119Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    12.8 ocular pigment deposit decrease - short term (<3 months)
119Risk Ratio (M-H, Fixed, 95% CI)0.37 [0.02, 8.01]

    12.9 nasal congestion - medium term (3-12 months)
125Risk Ratio (M-H, Fixed, 95% CI)0.23 [0.01, 5.12]

    12.10 dreams (unpleasant) - short term (<3 months)
119Risk Ratio (M-H, Fixed, 95% CI)3.3 [0.15, 72.08]

 13 Leaving the study early: 1. Due to adverse effects3104Risk Ratio (M-H, Fixed, 95% CI)0.21 [0.03, 1.33]

    13.1 short term (<3 months)
145Risk Ratio (M-H, Fixed, 95% CI)0.23 [0.02, 2.29]

    13.2 medium term (3-12 months)
259Risk Ratio (M-H, Fixed, 95% CI)0.19 [0.01, 3.75]

 14 Leaving the study early: 2. Any reason4Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    14.1 short term (<3 months)
390Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.30, 3.39]

    14.2 medium term (3-12 months)
266Risk Ratio (M-H, Fixed, 95% CI)0.35 [0.17, 0.72]

 
Comparison 2. PIMOZIDE versus PLACEBO: withdrawal study

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Global state: 1. Relapse - clinical diagnoses1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    1.1 medium term (3-12 months)
140Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.01, 7.72]

 2 Mental state: 1. Use of additional medication1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    2.1 medium term (3-12 months)
140Risk Ratio (M-H, Fixed, 95% CI)0.18 [0.06, 0.51]

 3 Adverse effects: 1. Death1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    3.1 medium term (3-12 months)
140Risk Ratio (M-H, Fixed, 95% CI)3.0 [0.13, 69.52]

 4 Leaving the study early: 1. Any reason1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    4.1 medium term (3-12 months)
140Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.07, 14.90]

 
Comparison 3. PIMOZIDE versus ANY ANTIPSYCHOTIC

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Service utilisation: 1. Hospital admission1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    1.1 medium term (3-12 months)
144Risk Ratio (M-H, Fixed, 95% CI)0.83 [0.35, 1.98]

 2 Global state: 1. Relapse - clinical diagnoses9Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    2.1 short term (<3 months)
260Risk Ratio (M-H, Fixed, 95% CI)1.67 [0.23, 11.87]

    2.2 medium term (3-12 months)
7227Risk Ratio (M-H, Fixed, 95% CI)0.82 [0.57, 1.17]

 3 Global state: 2. No improvement10Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    3.1 short term (<3 months)
5146Risk Ratio (M-H, Fixed, 95% CI)0.75 [0.48, 1.17]

    3.2 medium term (3-12 months)
6186Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.75, 1.31]

 4 Global state: 3. Average score - CGI (high = poor)1Mean Difference (IV, Fixed, 95% CI)Subtotals only

    4.1 short term (<3 months)
116Mean Difference (IV, Fixed, 95% CI)-0.15 [-1.25, 0.95]

 5 Mental state: 1. Average score - BPRS (high = poor)1Mean Difference (IV, Fixed, 95% CI)Subtotals only

    5.1 short term (<3 months)
116Mean Difference (IV, Fixed, 95% CI)-13.88 [-28.21, 0.45]

 6 Mental state: 2. Average score - MADRS (high = poor, skewed)Other dataNo numeric data

    6.1 short term (<3 months)
Other dataNo numeric data

 7 Mental state: 3. No improvement2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    7.1 short term (<3 months)
254Risk Ratio (M-H, Fixed, 95% CI)0.92 [0.43, 1.96]

    7.2 medium term (3-12 months)
123Risk Ratio (M-H, Fixed, 95% CI)1.09 [0.08, 15.41]

 8 Mental state: 4. Specific symptoms5Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    8.1 anxiety/tension - medium term (3-12 months)
3161Risk Ratio (M-H, Fixed, 95% CI)0.67 [0.20, 2.29]

    8.2 depression - short term (<3 months)
144Risk Ratio (M-H, Fixed, 95% CI)0.61 [0.30, 1.21]

    8.3 depression - medium term (3-12 months)
274Risk Ratio (M-H, Fixed, 95% CI)0.48 [0.26, 0.89]

    8.4 no improvement in psychomotor activity - short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)0.4 [0.10, 1.60]

    8.5 presence of first-rank symptoms - short term (<3 months)
144Risk Ratio (M-H, Fixed, 95% CI)0.56 [0.24, 1.29]

    8.6 presence of first-rank symptoms - medium term (3-12 months)
144Risk Ratio (M-H, Fixed, 95% CI)0.53 [0.25, 1.11]

 9 Mental state: 5. Use of additional medication3Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    9.1 short term (<3 months)
3114Risk Ratio (M-H, Fixed, 95% CI)1.07 [0.77, 1.49]

 10 Adverse effects: 1. Extrapyramidal adverse effects - specific13Risk Ratio (M-H, Random, 95% CI)Subtotals only

    10.1 akathisia - short term (<3 months)
4148Risk Ratio (M-H, Random, 95% CI)1.18 [0.79, 1.76]

    10.2 akathisia - medium term (3-12 months)
5219Risk Ratio (M-H, Random, 95% CI)1.17 [0.26, 5.22]

    10.3 acute dyskinesia - short term (<3 months)
129Risk Ratio (M-H, Random, 95% CI)2.68 [0.62, 11.64]

    10.4 dystonia - short term (<3 months)
242Risk Ratio (M-H, Random, 95% CI)0.78 [0.25, 2.48]

    10.5 dystonia - medium term (3-12 months)
250Risk Ratio (M-H, Random, 95% CI)1.30 [0.28, 6.00]

    10.6 gait disturbance - short term (<3 months)
259Risk Ratio (M-H, Random, 95% CI)1.18 [0.16, 8.62]

    10.7 gait disturbance - medium term (3-12 months)
138Risk Ratio (M-H, Random, 95% CI)0.25 [0.03, 2.04]

    10.8 hypersalivation - short term (<3 months)
3115Risk Ratio (M-H, Random, 95% CI)0.98 [0.31, 3.12]

    10.9 hypersalivation - medium term (3-12 months)
3102Risk Ratio (M-H, Random, 95% CI)0.88 [0.13, 6.15]

    10.10 involuntary movements - medium term (3-12 months)
187Risk Ratio (M-H, Random, 95% CI)0.34 [0.01, 8.14]

    10.11 Parkinsonism - short term (<3 months)
156Risk Ratio (M-H, Random, 95% CI)1.25 [0.37, 4.17]

    10.12 Parkinsonism - medium term (3-12 months)
2131Risk Ratio (M-H, Random, 95% CI)2.12 [0.28, 15.90]

    10.13 restlessness - short term (<3 months)
3101Risk Ratio (M-H, Random, 95% CI)1.20 [0.62, 2.34]

    10.14 restlessness - medium term (3-12 months)
3134Risk Ratio (M-H, Random, 95% CI)2.59 [0.41, 16.49]

    10.15 rigidity - short term (<3 months)
6186Risk Ratio (M-H, Random, 95% CI)1.21 [0.71, 2.05]

    10.16 rigidtiy - medium term (3-12 months)
5219Risk Ratio (M-H, Random, 95% CI)1.12 [0.24, 5.25]

    10.17 seizure - short term (<3 months)
129Risk Ratio (M-H, Random, 95% CI)0.36 [0.02, 8.07]

    10.18 tardive dyskinesia - short term (<3 months)
143Risk Ratio (M-H, Random, 95% CI)0.79 [0.05, 11.85]

    10.19 tardive dyskinesia - medium term (3-12 months)
4108Risk Ratio (M-H, Random, 95% CI)1.27 [0.73, 2.23]

    10.20 tremor - short term (<3 months)
6186Risk Ratio (M-H, Random, 95% CI)1.52 [0.97, 2.38]

    10.21 tremor - medium term (3-12 months)
4175Risk Ratio (M-H, Random, 95% CI)1.46 [0.68, 3.11]

    10.22 use of antiparkinsonian drugs - short term (<3 months)
267Risk Ratio (M-H, Random, 95% CI)1.46 [0.54, 3.94]

    10.23 use of antiparkinsonian drugs - by medium term (3-12 months)
3116Risk Ratio (M-H, Random, 95% CI)0.84 [0.41, 1.72]

 11 Adverse effects: 2. Extrapyramidal adverse effects: average score - EPS (high = poor, skewed)Other dataNo numeric data

    11.1 short term (<3 months)
Other dataNo numeric data

 12 Adverse effects: 3. Death1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    12.1 suicide - medium term (3-12 months)
151Risk Ratio (M-H, Fixed, 95% CI)0.32 [0.01, 7.53]

 13 Adverse effects: 4. Anticholinergic effects9Risk Ratio (M-H, Random, 95% CI)Subtotals only

    13.1 blurred vision - short term (<3 months)
4110Risk Ratio (M-H, Random, 95% CI)0.98 [0.46, 2.09]

    13.2 blurred vision - medium term (3-12 months)
5219Risk Ratio (M-H, Random, 95% CI)0.98 [0.45, 2.12]

    13.3 dry mouth - short term (<3 months)
5166Risk Ratio (M-H, Random, 95% CI)0.84 [0.54, 1.31]

    13.4 dry mouth - medium term (3-12 months)
4175Risk Ratio (M-H, Random, 95% CI)0.44 [0.23, 0.85]

    13.5 sweating - short term (<3 months)
285Risk Ratio (M-H, Random, 95% CI)2.16 [0.29, 15.96]

    13.6 urinary retention - short term (<3 months)
3121Risk Ratio (M-H, Random, 95% CI)1.26 [0.28, 5.67]

    13.7 urinary retention - medium term (3-12 months)
2125Risk Ratio (M-H, Random, 95% CI)0.26 [0.03, 2.25]

 14 Adverse effects: 5. Cardio-vascular effects10Risk Ratio (M-H, Random, 95% CI)Subtotals only

    14.1 chest pain - medium term (3-12 months)
130Risk Ratio (M-H, Random, 95% CI)0.33 [0.01, 7.58]

    14.2 dizziness - short term (<3 months)
5161Risk Ratio (M-H, Random, 95% CI)0.73 [0.26, 2.02]

    14.3 dizziness - medium term (3-12 months)
3161Risk Ratio (M-H, Random, 95% CI)1.08 [0.34, 3.43]

    14.4 ECG changes - short term (<3 months)
156Risk Ratio (M-H, Random, 95% CI)0.33 [0.04, 3.01]

    14.5 hypotension - short term (<3 months)
4143Risk Ratio (M-H, Random, 95% CI)1.13 [0.50, 2.59]

    14.6 hypotension - medium term (3-12 months)
143Risk Ratio (M-H, Random, 95% CI)0.32 [0.07, 1.46]

    14.7 hypotension (postural) - short term (<3 months)
242Risk Ratio (M-H, Random, 95% CI)0.33 [0.06, 1.92]

    14.8 hypertension - short term (<3 months)
156Risk Ratio (M-H, Random, 95% CI)1.0 [0.07, 15.21]

    14.9 hypertension - medium term (3-12 months)
187Risk Ratio (M-H, Random, 95% CI)0.51 [0.05, 5.44]

    14.10 palpitations - short term (<3 months)
278Risk Ratio (M-H, Random, 95% CI)1.28 [0.31, 5.28]

    14.11 substernal pain - medium term (3-12 months)
130Risk Ratio (M-H, Random, 95% CI)0.33 [0.01, 7.58]

    14.12 tachycardia - short term (<3 months)
129Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]

 15 Adverse effects: 6. Abnormal laboratory tests4Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    15.1 haematological - granulocytopenia - short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.02, 7.32]

    15.2 haematological - leukocytosis - short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.34, 2.93]

    15.3 haematological - leukocytosis - medium term (<3 months)
151Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.06, 14.55]

    15.4 haematological - macrocytosis - short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)5.0 [0.27, 92.62]

    15.5 haematological - any abnormal lab haematology - medium term (3-12 months)
130Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.31, 3.28]

    15.6 liver function - abnormal laboratory data - medium term (3-12 months)
130Risk Ratio (M-H, Fixed, 95% CI)0.2 [0.03, 1.51]

    15.7 renal function - albumin in urinalysis - short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)3.0 [0.14, 65.90]

    15.8 renal function - granular casts in urinalysis - short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)3.0 [0.14, 65.90]

    15.9 renal function - urea / nitrogen abnormal - medium term (3-12 months)
130Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

 16 Adverse effects: 7. Central nervous system effects12Risk Ratio (M-H, Random, 95% CI)Subtotals only

    16.1 drowsiness - short term (<3 months)
7226Risk Ratio (M-H, Random, 95% CI)0.77 [0.60, 1.00]

    16.2 drowsiness - medium term (3-12 months)
5229Risk Ratio (M-H, Random, 95% CI)0.61 [0.33, 1.11]

    16.3 excitement - short term (<3 months)
120Risk Ratio (M-H, Random, 95% CI)3.0 [0.14, 65.90]

    16.4 excitement - medium term (3-12 months)
187Risk Ratio (M-H, Random, 95% CI)0.68 [0.12, 3.88]

    16.5 faintness - medium term (3-12 months)
120Risk Ratio (M-H, Random, 95% CI)0.2 [0.01, 3.70]

    16.6 fatigue - short term (<3 months)
3101Risk Ratio (M-H, Random, 95% CI)1.07 [0.57, 2.00]

    16.7 headache - short term (<3 months)
4161Risk Ratio (M-H, Random, 95% CI)0.67 [0.21, 2.07]

    16.8 headache - medium term (3-12 months)
5219Risk Ratio (M-H, Random, 95% CI)2.14 [0.84, 5.45]

    16.9 insomnia - short term (<3 months)
3101Risk Ratio (M-H, Random, 95% CI)1.02 [0.67, 1.55]

    16.10 insomnia - medium term (3-12 months)
4181Risk Ratio (M-H, Random, 95% CI)0.92 [0.35, 2.39]

    16.11 memory defects - medium term (3-12 months)
130Risk Ratio (M-H, Random, 95% CI)0.33 [0.01, 7.58]

    16.12 restlessness - short term (<3 months)
156Risk Ratio (M-H, Random, 95% CI)0.88 [0.37, 2.09]

    16.13 restlessness - medium term (3-12 months)
295Risk Ratio (M-H, Random, 95% CI)2.46 [0.26, 22.89]

    16.14 sedation - short term (<3 months)
143Risk Ratio (M-H, Random, 95% CI)0.79 [0.47, 1.34]

    16.15 sedation - medium term (3-12 months)
3155Risk Ratio (M-H, Random, 95% CI)0.30 [0.12, 0.72]

 17 Adverse effects: 8. Dermatological effects7Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    17.1 acne vulgaris - short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)3.0 [0.14, 65.90]

    17.2 dermatitis - medium term (3-12 months)
130Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.07, 14.55]

    17.3 itching - medium term (3-12 months)
2131Risk Ratio (M-H, Fixed, 95% CI)0.61 [0.08, 4.46]

    17.4 rashes - short term (<3 months)
3145Risk Ratio (M-H, Fixed, 95% CI)1.75 [0.47, 6.53]

    17.5 rashes - medium term (3-12 months)
2125Risk Ratio (M-H, Fixed, 95% CI)0.38 [0.11, 1.31]

 18 Adverse effects: 9. Endocrine effects3Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    18.1 amenorrhoea - short term (<3 months)
156Risk Ratio (M-H, Fixed, 95% CI)5.0 [0.25, 99.67]

    18.2 amenorrhoea - medium term (3-12 months)
120Risk Ratio (M-H, Fixed, 95% CI)3.0 [0.14, 65.90]

    18.3 galactorrhoea - medium term (3-12 months)
130Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.01, 7.58]

 19 Adverse effects: 10. Gastrointestinal effects12Risk Ratio (M-H, Random, 95% CI)Subtotals only

    19.1 abdominal cramps - medium term (3-12 months)
3151Risk Ratio (M-H, Random, 95% CI)1.97 [0.35, 10.95]

    19.2 anal incontinence - short term (<3 months)
120Risk Ratio (M-H, Random, 95% CI)3.0 [0.14, 65.90]

    19.3 appetite decrease - short term (<3 months)
285Risk Ratio (M-H, Random, 95% CI)1.34 [0.22, 8.07]

    19.4 appetite decrease - medium term (3-12 months)
2107Risk Ratio (M-H, Random, 95% CI)1.05 [0.16, 6.84]

    19.5 appetite increase - short term (<3 months)
285Risk Ratio (M-H, Random, 95% CI)0.65 [0.15, 2.83]

    19.6 constipation - short term (<3 months)
5190Risk Ratio (M-H, Random, 95% CI)0.64 [0.31, 1.30]

    19.7 constipation - medium term (3-12 months)
4132Risk Ratio (M-H, Random, 95% CI)0.75 [0.30, 1.88]

    19.8 diarrhoea - short term (<3 months)
276Risk Ratio (M-H, Random, 95% CI)1.02 [0.11, 9.33]

    19.9 diarrhoea - medium term (3-12 months)
130Risk Ratio (M-H, Random, 95% CI)3.0 [0.13, 68.26]

    19.10 nausea and vomiting - short term (<3 months)
276Risk Ratio (M-H, Random, 95% CI)0.28 [0.05, 1.61]

    19.11 nausea - short term (<3 months)
116Risk Ratio (M-H, Random, 95% CI)5.0 [0.28, 90.18]

    19.12 nausea - medium term (3-12 months)
5203Risk Ratio (M-H, Random, 95% CI)0.64 [0.24, 1.69]

    19.13 vomiting - short term (<3 months)
259Risk Ratio (M-H, Random, 95% CI)3.0 [0.14, 64.26]

    19.14 vomiting - medium term (3-12 months)
2125Risk Ratio (M-H, Random, 95% CI)0.77 [0.08, 7.40]

    19.15 acute cholecystitis - short term (<3 months)
160Risk Ratio (M-H, Random, 95% CI)2.81 [0.12, 66.40]

 20 Adverse effects: 11. Genitourinary effects3Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    20.1 anorgasmia - medium term (3-12 months)
130Risk Ratio (M-H, Fixed, 95% CI)3.0 [0.13, 68.26]

    20.2 nocturia - medium term (3-12 months)
187Risk Ratio (M-H, Fixed, 95% CI)0.34 [0.01, 8.14]

    20.3 dysuria - short term (<3 months)
156Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.01, 7.85]

 21 Adverse effects: 12. Other11Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    21.1 increased thirst - short term (<3 months)
129Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    21.2 nasal congestion - medium term (3-12 months)
130Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]

    21.3 intensified symptoms - short term (<3 months)
2116Risk Ratio (M-H, Fixed, 95% CI)0.48 [0.09, 2.50]

    21.4 intensified symptoms - medium term (3-12 months)
292Risk Ratio (M-H, Fixed, 95% CI)0.59 [0.27, 1.28]

    21.5 weight loss - short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.04, 2.69]

    21.6 weight loss - medium term (3-12 months)
3161Risk Ratio (M-H, Fixed, 95% CI)1.38 [0.48, 3.98]

    21.7 weight gain - short term (<3 months)
143Risk Ratio (M-H, Fixed, 95% CI)1.58 [0.16, 16.17]

    21.8 weight gain - medium term (3-12 months)
4175Risk Ratio (M-H, Fixed, 95% CI)1.55 [0.56, 4.33]

 22 Social functioning: 1. Average score - SBAS (high = poor, skewed)Other dataNo numeric data

    22.1 medium term (3-12 months)
Other dataNo numeric data

 23 Leaving the study early: 1. Due to relapse4Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    23.1 short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)3.0 [0.14, 65.90]

    23.2 medium term (3-12 months)
376Risk Ratio (M-H, Fixed, 95% CI)0.88 [0.28, 2.78]

 24 Leaving the study early: 2. Due to adverse effects14Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    24.1 short term (<3 months)
7258Risk Ratio (M-H, Fixed, 95% CI)0.80 [0.32, 2.01]

    24.2 medium term (3-12 months)
7252Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.57, 1.72]

 25 Leaving the study early: 3. Any reason25Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    25.1 short term (<3 months)
11368Risk Ratio (M-H, Fixed, 95% CI)1.23 [0.63, 2.40]

    25.2 medium term (3-12 months)
14523Risk Ratio (M-H, Fixed, 95% CI)0.87 [0.69, 1.09]

    25.3 long term (>12 months)
162Risk Ratio (M-H, Fixed, 95% CI)1.0 [0.60, 1.67]

 
Comparison 4. PIMOZIDE + ANY ANTIPSYCHOTIC versus ANY ANTIPSYCHOTIC

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Global state: 1. Relapse - clinical diagnoses1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    1.1 medium term (3-12 months)
169Risk Ratio (M-H, Fixed, 95% CI)0.28 [0.15, 0.50]

 2 Leaving the study early: 1. Any reason1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    2.1 medium term (3-12 months)
169Risk Ratio (M-H, Fixed, 95% CI)0.56 [0.42, 0.74]

 
Comparison 5. PIMOZIDE + ANY ANTIPSYCHOTIC versus ANY ANTIPSYCHOTIC + PLACEBO

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Global state: 1. Average score - CGI (high = poor, skewed)Other dataNo numeric data

    1.2 medium term (3-12 months)
Other dataNo numeric data

 2 Global state: 2. Average score - CGI-S (high = poor)1Mean Difference (IV, Fixed, 95% CI)Subtotals only

    2.1 short term (<3 months)
128Mean Difference (IV, Fixed, 95% CI)0.0 [-0.49, 0.49]

 3 Global state: 3. Average score - CGI-I (high = poor)1Mean Difference (IV, Fixed, 95% CI)Subtotals only

    3.1 short term (<3 months)
128Mean Difference (IV, Fixed, 95% CI)0.60 [0.01, 1.19]

 4 Mental state: 1. Average score - BPRS (high = poor)1Mean Difference (IV, Fixed, 95% CI)Subtotals only

    4.1 short term (<3 months)
128Mean Difference (IV, Fixed, 95% CI)1.70 [0.14, 3.26]

 5 Mental state: 2. Average score - BPRS psychosis subscale (high = poor)1Mean Difference (IV, Fixed, 95% CI)Subtotals only

    5.1 short term (<3 months)
128Mean Difference (IV, Fixed, 95% CI)1.70 [0.96, 2.44]

 6 Mental state: 3. Average score - PANSS (high = poor, skewed)Other dataNo numeric data

    6.1 medium term (3-12 months)
Other dataNo numeric data

 7 Mental state: 4. Average score - SANS (high = poor)1Mean Difference (IV, Fixed, 95% CI)Subtotals only

    7.1 short term (<3 months)
128Mean Difference (IV, Fixed, 95% CI)1.0 [-3.37, 5.37]

 8 Adverse effects: 1. Extrapyramidal adverse effects - specific1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    8.1 hypersalivation - medium term (3-12 months)
153Risk Ratio (M-H, Fixed, 95% CI)2.99 [0.89, 10.04]

 9 Adverse effects: 2. Extrapyramidal adverse effects: average score - ESRS (high = poor, skewed)Other dataNo numeric data

    9.1 parkinsonism - medium term (3-12 months)
Other dataNo numeric data

    9.2 dystonia - medium term (3-12 months)
Other dataNo numeric data

    9.3 dyskinesia - medium term (3-12 months)
Other dataNo numeric data

 10 Adverse effects: 3. Cardiovascular effects1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    10.1 bigeminy - medium term (3-12 months)
153Risk Ratio (M-H, Fixed, 95% CI)0.37 [0.02, 8.73]

    10.2 hypotension - medium term (3-12 months)
153Risk Ratio (M-H, Fixed, 95% CI)0.37 [0.02, 8.73]

 11 Adverse effects: 4. Cardiovascular effects - average score (high = poor)1Mean Difference (IV, Fixed, 95% CI)Subtotals only

    11.1 QTc (ms) - short term (<3 months)
128Mean Difference (IV, Fixed, 95% CI)7.40 [3.62, 11.18]

 12 Adverse effects: 5. Cardiovascular effects - average score (high = poor, skewed)Other dataNo numeric data

    12.1 diastolic blood pressure (mm Hg) - medium term (3-12 months)
Other dataNo numeric data

    12.2 heart rate (bpm) - medium term (3-12 months)
Other dataNo numeric data

    12.3 QTc (ms) - medium term (3-12 months)
Other dataNo numeric data

    12.4 systolic blood pressure (mm Hg) - medium term (3-12 months)
Other dataNo numeric data

 13 Adverse effects: 6. Abnormal laboratory tests - average score (skewed)Other dataNo numeric data

    13.1 absolute neutrophil count (count/mm 3) - medium term (3-12 months)
Other dataNo numeric data

    13.2 plasma glucose level (mg/dL) - medium term (3-12 months)
Other dataNo numeric data

    13.3 total cholesterol level (mg/dL) - medium term (3-12 months)
Other dataNo numeric data

    13.4 triglyceride level (mg/dL) - medium term (3-12 months)
Other dataNo numeric data

 14 Adverse effects: 7. Other1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    14.1 intensified symptoms - medium term (3-12 months)
153Risk Ratio (M-H, Fixed, 95% CI)0.22 [0.01, 4.44]

 15 Quality of life: 1. Average score - SLOF (high = better, skewed)Other dataNo numeric data

    15.1 physical function (SLOF) - medium term (3-12 months)
Other dataNo numeric data

    15.2 personal care (SLOF) - medium term (3-12 months)
Other dataNo numeric data

    15.3 interpersonal relationships (SLOF) - medium term (3-12 months)
Other dataNo numeric data

    15.4 social acceptability (SLOF) - medium term (3-12 months)
Other dataNo numeric data

    15.5 work skills (SLOF) - medium term (3-12 months)
Other dataNo numeric data

 16 Leaving the study early: 1. Due to adverse effects281Risk Ratio (M-H, Fixed, 95% CI)0.85 [0.22, 3.25]

    16.1 short term (<3 months)
128Risk Ratio (M-H, Fixed, 95% CI)3.0 [0.13, 67.91]

    16.2 medium term (3-12 months)
153Risk Ratio (M-H, Fixed, 95% CI)0.56 [0.11, 2.80]

 17 Leaving the study early: 2. Any reason281Risk Ratio (M-H, Fixed, 95% CI)1.09 [0.35, 3.33]

    17.1 short term (<3 months)
128Risk Ratio (M-H, Fixed, 95% CI)5.0 [0.26, 95.61]

    17.2 medium term (3-12 months)
153Risk Ratio (M-H, Fixed, 95% CI)0.67 [0.18, 2.53]

 
Comparison 6. PIMOZIDE + ANY ANTIPSYCHOTIC versus ANY ANTIPSYCHOTIC + ANY ANTIPSYCHOTIC

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Global state: 1. No improvement1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    1.1 short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)5.0 [0.70, 35.50]

 2 Global state: 2. Average score - CGI-S (high = poor)1Mean Difference (IV, Fixed, 95% CI)Subtotals only

    2.1 short term (<3 months)
120Mean Difference (IV, Fixed, 95% CI)2.5 [1.30, 3.70]

 3 Global state: 3. Average score - CGI-I (high = poor)1Mean Difference (IV, Fixed, 95% CI)Subtotals only

    3.1 short term (<3 months)
120Mean Difference (IV, Fixed, 95% CI)1.50 [0.93, 2.07]

 4 Mental state: 1. Average score - SANS (high = poor)1Mean Difference (IV, Fixed, 95% CI)Subtotals only

    4.1 short term (<3 months)
120Mean Difference (IV, Fixed, 95% CI)24.0 [5.98, 42.02]

 5 Mental state: 2. Average score - SAPS (high = poor, skew)Other dataNo numeric data

    5.1 short term (<3 months)
Other dataNo numeric data

 6 Adverse effects: 1. Extrapyramidal adverse effects - average score (high = poor, skew)Other dataNo numeric data

    6.1 akathisia - short term (<3 months)
Other dataNo numeric data

    6.2 akinesia - short term (<3 months)
Other dataNo numeric data

    6.3 EPS symptoms - short term (<3 months)
Other dataNo numeric data

    6.4 rigidity - short term (<3 months)
Other dataNo numeric data

    6.5 tremor - short term (<3 months)
Other dataNo numeric data

 7 Adverse effects: 2. Anticholinergenic effects - average score (high = poor, skew)Other dataNo numeric data

    7.1 blurred vision - short term (<3 months)
Other dataNo numeric data

    7.2 dry mouth - short term (<3 months)
Other dataNo numeric data

 8 Adverse effects: 3. Cardio-vascular effects - average score (high = poor, skew)Other dataNo numeric data

    8.1 dizziness - short term (<3 months)
Other dataNo numeric data

    8.2 hypotension - short term (<3 months)
Other dataNo numeric data

    8.3 tachycardia - short term (<3 months)
Other dataNo numeric data

 9 Adverse effects: 4. Autonomic effects - average score (high = poor, skew)Other dataNo numeric data

    9.1 autonomic effects - short term (<3 months)
Other dataNo numeric data

 10 Adverse effects: 5. Central nervous system effects - average score (high = poor, skew)Other dataNo numeric data

    10.1 drowsiness - short term (<3 months)
Other dataNo numeric data

    10.2 headache - short term (<3 months)
Other dataNo numeric data

    10.3 insomnia - short term (<3 months)
Other dataNo numeric data

 11 Adverse effects: 6. Endocrine effects - average score (high = poor, skew)Other dataNo numeric data

    11.1 amenorrhoea - short term (<3 months)
Other dataNo numeric data

    11.2 endocrine system effects - short term (<3 months)
Other dataNo numeric data

    11.3 galactorrhoea - short term (<3 months)
Other dataNo numeric data

 12 Adverse effects: 7. Gastrointestinal effects - average score (high = poor, skew)Other dataNo numeric data

    12.1 constipation - short term (<3 months)
Other dataNo numeric data

    12.2 nausea - short term (<3 months)
Other dataNo numeric data

    12.3 weight gain - short term (<3 months)
Other dataNo numeric data

    12.4 weight loss - short term (<3 months)
Other dataNo numeric data

 13 Adverse effects: 8. Gastrointestinal effects1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    13.1 constipation - short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)3.0 [0.14, 65.90]

 14 Adverse effects: 9. Other1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    14.1 mammary tension - short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)5.0 [0.70, 35.50]

 15 Adverse effects: 10. Other - average score (high = poor, skew)Other dataNo numeric data

    15.1 gynaecomastia - short term (<3 months)
Other dataNo numeric data

 16 Leaving the study early: 1. Due to adverse effects1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    16.1 short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.02, 7.32]

 17 Leaving the study early: 2. Any reason1Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    17.1 short term (<3 months)
120Risk Ratio (M-H, Fixed, 95% CI)0.33 [0.02, 7.32]

 
Comparison 7. Sensitivity Analysis: PIMOZIDE vs ANY ANTIPSYCHOTIC - 1. risk of bias

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Mental state: 1. No improvement2Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    1.1 unclear
124Risk Ratio (M-H, Fixed, 95% CI)0.75 [0.21, 2.66]

    1.2 high risk
130Risk Ratio (M-H, Fixed, 95% CI)1.05 [0.41, 2.70]

 2 Extrapyramidal adverse effects: 1. Specific - Parkinsonism3187Risk Ratio (M-H, Fixed, 95% CI)1.55 [0.56, 4.26]

    2.1 unclear
2143Risk Ratio (M-H, Fixed, 95% CI)1.21 [0.40, 3.64]

    2.2 high risk
144Risk Ratio (M-H, Fixed, 95% CI)5.0 [0.25, 98.52]

 

Appendices

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms
 

Appendix 1. Previous methods

 

Criteria for considering studies for this review  

 

Types of studies  

We included all relevant randomised controlled trials. Where the allocation method was unclear, we included these studies only when they did not change the results of the review. We excluded quasi-randomised studies such as those allocating by alternate days of the week.

 

Appendix 2. Previous search strategies

 
1. Electronic searches
 
1.1. Update (2005)

We searched the Cochrane Schizophrenia Group Trials Register (July 2005) using the phrase:

[(pimozide or orap or antalon or opiran or pirium in REFERENCE title, abstract and index fields) OR (pimozide or orap or antalon or opiran or pirium) in STUDY intervention field]

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

 
1.2. Original search (2002)

The Cochrane Schizophrenia Group Trials Register (September 2002) was searched using the phrase:

 [(*pimozide* or * orap* or *antalon* or *opiran* or *pirium* in title, abstract or index terms of REFERENCE) or  (pimozide / R.6238 in interventions of STUDY)]

The Cochrane Schizophrenia Group Trials Register is based on regular searches of BIOSIS Inside; CENTRAL; CINAHL; EMBASE; MEDLINE and PsycINFO; the handsearching of relevant journals and conference proceedings and searches of several key grey literature sources. A full description is given in the Group’s module.

 
2. Details of previous searches:
 
2.1. Electronic searching

We identified relevant randomised trials by searching the following electronic databases:

2.1.1. We searched Biological Abstracts (January 1982 to April 1999) using the Cochrane Schizophrenia Group's phrase for randomised controlled trials (see Group search strategy) combined with:

[and PIMOZIDE or ORAP or ANTALON or OPIRAN or PIRIUM]

2.1.2. We searched CINAHL (January 1982 to April 1999) using the Cochrane Schizophrenia Group's phrase for randomised controlled trials (see Group search strategy) combined with:

[and PIMOZIDE or ORAP or ANTALON or OPIRAN or PIRIUM]

2.1.3. We searched the Cochrane Schizophrenia Group's Register (April 1999) using the phrase:

[PIMOZIDE or ORAP or ANTALON or OPIRAN or PIRIUM or #42=1]

#42 is the field in this register in which codes for each intervention are stored, and 1 is the code for pimozide.

2.1.4. We searched The Cochrane Library (Issue 1, 1998) using the phrase:

[PIMOZIDE or ORAP or ANTALON or OPIRAN or PIRIUM or PIMOZIDE*:ME]

2.1.5. We searched EMBASE (January 1980 to April 1999) using the Cochrane Schizophrenia Group's phrase for randomised controlled trials (see Group search strategy) combined with:

[and (PIMOZIDE or ORAP or ANTALON or OPIRAN or PIRIUM or explode "PIMOZIDE"/ all subheadings)]

2.1.6. We searched MEDLINE (January 1966 to April 1999) using the Cochrane Schizophrenia Group's phrase for randomised controlled trials (see Group search strategy) combined with:

[and ((explode "PIMOZIDE"/ all subheadings) or PIMOZIDE or ORAP or ANTALON or OPIRAN or PIRIUM)]

2.1.7. We searched PsycLIT (January 1887 to April 1999) using the Cochrane Schizophrenia Group's phrase for randomised

 
2.2 Reference searching

We inspected the references of all identified studies to find more studies.

 
2.3 Personal contact

We contacted the first author of each included study to ask for information regarding unpublished trials.

 
2.4 Drug company

We contacted the manufacturers of pimozide (Janssen-Cilag) to ask for additional data.

 
2.5. Handsearching

It was originally planned that high-yield journals would be identified by electronic searches and if available and not already handsearched would be subjected to page-by-page inspection.

 

Appendix 3. Previous data collection and analysis

[For definitions of terms used in this, and other sections, please refer to the Glossary]

 
1. Selection of trials

Material downloaded from electronic sources included details of author, institution or journal of publication.

For the original publication, the principal review author (AS) inspected all reports. These were then re-inspected by TM to ensure reliable selection. We resolved any disagreement by discussion, and when doubt contiinued, we acquired the full article for further inspection. Once the full articles were obtained, we (AS and TM) decided whether the studies met the review criteria. If disagreement could not be resolved by discussion, we sought further information, and these trials were added to the list of those awaiting assessment. For the 2005 update, JR independently selected studies identified through the 2005 search strategy.

 
2. Assessment of methodological quality

We assessed the methodological quality of included trials in this review using the criteria described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2005) and the Jadad Scale (Jadad 1996). The former is based on evidence of a strong relationship between allocation concealment and direction of effect (Schulz 1995). The categories are defined below.

A. Low risk of bias (adequate allocation concealment).
B. Moderate risk of bias (some doubt about the results).
C. High risk of bias (inadequate allocation concealment). For the purpose of the analysis in this review, trials were included if they met the Cochrane Handbook for Systematic Reviews of Interventions criterion A or B.

The Jadad Scale measures a wider range of factors that can impact the quality of a trial. The scale includes three items.
1. Was the study described as randomised?
2. Was the study described as double-blind?
3. Was there a description of withdrawals and people leaving the study early?

Each item receives one point if the answer is positive. In addition, a point can be deducted if the randomisation or the blinding/masking procedures described are inadequate. For this review, we used a cut-off of two points on the Jadad Scale to check the assessment made in accordace with the Cochrane Handbook for Systematic Reviews of Interventions criteria. However, we did not use the Jadad Scale to exclude trials.

 
3. Data collection

AS independently extracted data from selected trials, while TM separately re-extracted information from two different samples (10%). When disputes arose, we attempted to resolve these by discussion. When this was not possible and further information was necessary to resolve the dilemma, data were not entered, and we added the trial to the list of those awaiting assessment. For the 2005 update, JR independently extracted data. When doubts arose as to the appropriateness of a study, this was discussed with the Cochrane Schizophrenia Group's Co-ordinating Editor.

 
4. Data synthesis
 
4.1 Data types

We assessed outcomes using continuous (e.g. changes on a behaviour scale), categorical (e.g. one of three categories on a behaviour scale, such as 'little change', 'moderate change' or 'much change') or dichotomous (e.g. 'no important changes' or 'important changes' in a person's behaviour) measures. Currently RevMan does not support categorical data, so we were unable to analysis this.

 
4.2 Incomplete data

We did not include trial outcomes if more than 40% of participants were not reported in the final analysis.

 
4.3 Dichotomous-yes/no-data

We carried out an intention-to-treat analysis. On the condition that more than 60% of people completed the study, all participants allocated to the intervention were counted, whether they completed the follow-up or not. It was assumed that those who dropped out had the negative outcome, with the exception of death. Where possible, efforts were made to convert outcome measures to dichotomous data. This can be done by identifying cut-off points on rating scales and dividing participants accordingly into 'clinically improved' and 'not clinically improved'. If the authors of a study had used a predefined cut-off point for determining clinical effectiveness, this was used by the review authors where appropriate. Otherwise it was generally assumed that if there a 50% reduction had been noted in a scale-derived score, this could be considered as a clinically significant response. Similarly, a rating of 'at least much improved' according to the Clinical Global Impression Scale (Guy 1976) was considered as a clinically significant response.

The relative risk (RR) and its 95% confidence interval (CI) were calculated on the basis of the random-effects model, as this takes into account any differences between studies, even if no statistically significant heterogeneity is observed. It has been shown that RR is more intuitive (Boiseel 1999) than odds ratios, which tend to be interpreted as RR by clinicians (Deekes 2000). This misinterpretation then leads to an overestimate of the impression of the effect. We inspected data to see whether an analysis using a fixed-effect model made any substantive difference in outcomes that were not statistically significantly heterogeneous. When the overall results were significant, we calculated the number needed to treat for an additional beneficial outcome (NNTB) and the number needed to treat for an additional harmful outcome (NNTH) as the inverse of the risk difference.

 
4.4 Continuous data

4.4.1 Normally distributed data: Continuous data on clinical and social outcomes often are not normally distributed. To avoid the pitfall of applying parametric tests to non-parametric data, we applied the following standards to all 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 the finite number zero, the standard deviation, when multiplied by two, was less than the mean (as otherwise, the mean is unlikely to be an appropriate measure of the centre of the distribution) (Altman 1996); (c) if a scale started from a positive value (such as PANSS, which can have values from 30 to 210), the calculation described above was modified to take the scale starting point into account. In these cases, skew is present if 2 SD > (S - Smin), where S is the mean score and Smin is the minimum score. Endpoint scores on scales often have finite starting and ending points, and these rules can be applied to them. When continuous data are presented on a scale that includes a possibility of negative values (such as change on a scale), it is difficult to tell whether or not data are non-normally distributed (skewed). Skewed data from studies of fewer than 200 participants would have been entered into additional tables rather than into an analysis. Skewed data pose less of a problem when one is looking at means if the sample size is large and would have been entered into a synthesis.

For change data (endpoint minus baseline), the situation is even more problematic. In the absence of individual participant data, it is impossible to know whether data are skewed, although this is likely. After consulting the ALLSTAT electronic statistics mailing list, we presented change data in MetaView to summarise available information. In doing this, it was assumed that data were not skewed or that the analyses could cope with the unknown degree of skew. Without individual participant data, it is impossible to test this assumption. When both change and endpoint data were available for the same outcome category, only endpoint data were presented. We acknowledge that by doing this many of the published change data were excluded, but we argue that endpoint data are more clinically relevant and that if change data were to be presented along with endpoint data, this would be given undeserved equal prominence. We are contacting authors of studies reporting only change data for endpoint figures. We reported non-normally distributed data in the 'other data types' tables.

4.4.2 Rating scales: A wide range of instruments are available to measure mental health outcomes. These instruments vary in quality, and many are not valid, or even ad hoc. For outcome instruments, some minimum standards have to be set. It has been shown that the use of rating scales that have not been described in a peer-reviewed journal (Marshall 2000) is associated with bias; therefore the results of such scales were excluded. Furthermore, we stipulated that the instrument shouldbe a self-report or should be completed by an independent rater or relative (not the therapist), and that the instrument could be considered a global assessment of an area of functioning. However, as it was expected that therapists would frequently also be the raters, such data were included but were commented on as 'prone to bias'.

Whenever possible, we took the opportunity to make direct comparisons between trials that used the same measurement instrument to quantify specific outcomes. When continuous data were presented from different scales in rating the same effect, both sets of data were presented, and the general direction of effect was inspected.

4.4.3 Summary statistic
For continuous outcomes, we estimated a weighted mean difference (WMD) between groups, again based on the random-effects model, as this takes into account any differences between studies, even if no statistically significant heterogeneity is noted.

 
4.5 Cluster trials

Studies increasingly employ 'cluster randomisation' (such as randomisation by clinician or practice), but analysis and pooling of clustered data pose problems. First, authors often fail to account for intraclass 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).

When clustering was not accounted for in primary studies, we presented the data in a table, with an (*) 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 intraclass correlation coefficients of their clustered data and to adjust for these using accepted methods (Gulliford 1999). Where clustering has been incorporated into the analysis of primary studies, we will 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 intraclass correlation coefficient (ICC) [Design effect = 1 + (m - 1) * ICC] (Donner 2002). If the ICC was not reported, it was assumed to be 0.1 (Ukoumunne 1999).

If cluster studies had been appropriately analysed, with intraclass correlation coefficients and relevant data documented in the report taken into account, synthesis with other studies would have been possible using the generic inverse variance technique.

 
5. Investigation for heterogeneity

First, we considered all included studies within any comparison to judge clinical heterogeneity. Then graphs were visually inspected to investigate the possibility of statistical heterogeneity. This was supplemented using, primarily, the I2 statistic. This provides an estimate of the percentage of variability due to heterogeneity rather than chance alone. When the I2 estimate was greater than or equal to 75%, we interpreted this as indicating the presence of high levels of heterogeneity (Higgins 2003). If inconsistency was high, we did not summate data but presented the data separately and investigated the reasons for heterogeneity.

 
6. Addressing publication bias

We entered data from all identified and selected trials into a funnel graph (trial effect vs trial size) in an attempt to investigate the likelihood of overt publication bias.

 
7. Sensitivity analyses

If available, results for delusional disorder will be compared with results for other psychotic diagnoses with regard to primary outcomes.

 
8. General

When possible, we entered data in such a way that the area to the left of the line of no effect indicated a favourable outcome for pimozide.

 

What's new

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms

Last assessed as up-to-date: 17 May 2007.


DateEventDescription

11 September 2013New citation required and conclusions have changedMinor changes to conclusions after results of 2013 search added to review

28 February 2013New search has been performedResults of 2011 and 2013 update search added. Six previously included studies have been excluded, and three new studies have been identified and included, bringing the total number of included studies in the review to 32; new comparisons and outcomes added with new data, resulting in minor changes to conclusions.



 

History

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms

Protocol first published: Issue 3, 1999
Review first published: Issue 1, 2000


DateEventDescription

5 August 2009New search has been performedMinor update.

23 April 2008New search has been performedRepublished.

23 April 2008AmendedConverted to new review format.

17 May 2007New citation required and conclusions have changedSubstantive amendment.

17 May 2007AmendedAmended.

25 February 2003AmendedEdited.

1 June 2000AmendedSubstantive amendment.

26 October 1999New search has been performedReview first published.



 

Contributions of authors

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms

Meghana Mothi - (update 2012) selected studies, extracted data and helped with report writing.

Stephanie Sampson - (update 2012) selected studies, extracted data and helped with report writing.

 

Declarations of interest

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms

None known.

 

Sources of support

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms
 

Internal sources

  • St. Patrick's Hospital, Dublin, Ireland.

 

External sources

  • National Institute for Health Research (NIHR), UK.
    Cochrane Collaboration Programme Grant 2011

 

Differences between protocol and review

  1. Top of page
  2. Summary of findings    [Explanations]
  3. Background
  4. Objectives
  5. Methods
  6. Results
  7. Discussion
  8. Authors' conclusions
  9. Acknowledgements
  10. Data and analyses
  11. Appendices
  12. What's new
  13. History
  14. Contributions of authors
  15. Declarations of interest
  16. Sources of support
  17. Differences between protocol and review
  18. Index terms

See  Table 2 on differences between protocol and review, which lists changes as 'minor' or 'major' and details differences between the previously published review and this 2013 updated version.

* Indicates the major publication for the study

References

References to studies included in this review

  1. Top of page
  2. AbstractRésumé scientifique
  3. Summary of findings
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' conclusions
  10. Acknowledgements
  11. Data and analyses
  12. Appendices
  13. What's new
  14. History
  15. Contributions of authors
  16. Declarations of interest
  17. Sources of support
  18. Differences between protocol and review
  19. Characteristics of studies
  20. References to studies included in this review
  21. References to studies excluded from this review
  22. References to studies awaiting assessment
  23. Additional references
  24. References to other published versions of this review
Abuzzahab 1980* {published data only}
  • Abuzzahab FS Sr. Evaluations of social functioning in a 3-year double-blind investigation of pimozide versus fluphenazine in chronic schizophrenia. Psychopharmacology Bulletin 1977;13:71-3.
  • Abuzzahab FS Sr, Zimmermann RL. A three-year double-blind investigation of pimozide versus fluphenazine in chronic schizophrenia. Psychopharmacology Bulletin 1976;12:26-7.
  • Abuzzahab FS, Zimmermann RL. Factors determining patient tenure on a 3-year double-blind investigation of pimozide versus fluphenazine HCl. Advances in Biochemistry and Psychopharmacology 1980;24:547-50.
Amin 1977 {published data only}
  • Amin MM, Ban TA, Lehmann HE. A standard (trifluoperazine) controlled clinical study with pimozide in the maintenance treatment of schizophrenic patients. Psychopharmacology Bulletin 1977;13:15-7.
Andersen 1972 {published data only}
  • Andersen K, d'Elia G, Hallberg B, Perris C, Rapp W, Roman G. A controlled trial of pimozide and trifluoperazine in chronic schizophrenic syndromes. Acta Psychiatrica Scandinavica Supplementum 1974;249:43-64.
Anumonye 1976 {published data only}
  • Anumonye A, Onibuwe-Johnson T, Marinho AA. Clinical trial of pimozide. West African Journal of Pharmacology and Drug Research 1976;3:17-24.
Barnes 1983 {published data only}
  • Barnes TR, Milavic G, Curson DA, Platt SD. Use of the Social Behaviour Assessment Schedule (SBAS) in a trial of maintenance antipsychotic therapy in schizophrenic outpatients: pimozide versus fluphenazine. Social Psychiatry 1983;18:193-9.
  • Barnes TR, Platt SD. E-mail correspondence. E-mail correspondence June 2011.
Chouinard 1970 {published data only}
  • Chouinard G, Lehmann HE, Ban TA. Pimozide in the treatment of chronic schizophrenic patients. Current Therapeutic Research and Clinical Experience 1970;12:598-603.
Chouinard 1982 {published data only}
Claghorn 1974 {published data only}
Clark 1975* {published data only}
  • Clark ML. Pimozide vs thioridazine vs placebo. Early Clinical Drug Evaluation Unit Reports 1973; Vol. 9.
  • Clark ML, Huber WK, Hill D, Wood F, Costiloe JP. Pimozide in chronic schizophrenic outpatients. Diseases of the Nervous System 1975;36:137-41.
Denijs 1973 {published data only}
  • Denijs EL, Vereecken JL. Pimozide (orap, R 6238) in residual schizophrenia: a clinical evaluation with long-term double-blind follow-up. Psychiatria Neurologia Neurochirurgia 1973;76:47-59.
De Ronchi 1996 {published data only}
  • De Ronchi D. Levosulpiride vs pimozide in negative symptoms of schizophrenia. Proceedings of the Xth World Congress of Psychiatry; August 23-28, 1996; Madrid, Spain. 1996.
  • De Ronchi D, Belelli G, Ruggeri M, Linari F, Tonti L, Volterra V. Levosulpiride vs pimozide in schizophrenic patients: efficacy and adverse events [Levosulpiride vs pimozide in pazienti schizofrenici: efficacia ed eventi avversi]. Neurologia Psichiatria Scienze Umane 1997;17(6):859-73. [EMBASE: 1998134063]
  • De Ronchi D, Ruggeri M, Balelli G, Volterra V. Levosulpiride vs pimozide in negative symptoms of schizophrenia. Proceedings of the 8th ECNP (European College of Neuropsychopharmacology) Congress; September 30-October 4, 1995; Venice, Italy. 1995.
  • De Ronchi D, Ruggeri M, Belelli G, Volterra V. Levosulpiride versus pimozide in negative symptoms of schizophrenia. Current Therapeutic Research - Clinical and Experimental 1996;57:797-810.
Donlon 1977 {published data only}
  • Donlon PT, Swaback DO, Osborne ML. Pimozide versus fluphenazine in ambulatory schizophrenics: a 12-month comparison study. Diseases of the Nervous System 1977;38:119-23.
Falloon 1978 {published data only}
  • Falloon I, Watt DC, Shepherd M. A comparative controlled trial of pimozide and fluphenazine decanoate in the continuation therapy of schizophrenia. Psychological Medicine 1978;8:59-70.
  • Falloon I, Watt DC, Shepherd M. The social outcome of patients in a trial of long-term continuation therapy. Psychological Medicine 1978;8:265-74.
  • Shepherd M. Medico-social evaluation of the long term pharmacotherapy of schizophrenia: comparaive study of fluphenazine and pimozide. Progressive Neuro-Psychopharmacology 1979;3:383-9.
Friedman 2011 {published data only}
  • Friedman JI, Lindenmayer J-P, Alcantara F, Bowler S, Parak M, White L, et al. Pimozide augmentation of clozapine inpatients with schizophrenia and schizoaffective disorder unresponsive to clozapine monotherapy. Neuropsychopharmacology 2011;36(6):1289-95. [MEDLINE: BIOSIS:PREV201100289820]
Gowardman 1973 {published data only}
Gross 1974* {published data only}
  • Gross-HS. A double-blind comparison of once-a-day pimozide, trifluoperazine, and placebo in the maintenance care of chronic schizophrenic outpatients. Current Therapeutic Research and Clinical Experience 1974;16:696-705.
Gunduz-Bruce 2013 {published and unpublished data}
  • Gunduz-Bruce H. Efficacy of pimozide augmentation for clozapine partial response. Clinicaltrials.gov. [NCT00374244]
  • Gunduz-Bruce H, Oliver S, Gueorguieva R, Forselius-Bielen K, D'Souza DC, Zimolo Z, et al. Efficacy of pimozide augmentation for clozapine partial responders with schizophrenia. Schizophrenia Research 2013;143:344-7.
Haas 1982 {published data only}
Huber 1971 {published data only}
  • Huber W, Serafetinides EA, Colmore JP, Clark M. Pimozide in chronic schizophrenic patients. Journal of Clinical Pharmacology and New Drugs 1971;11:304-9.
Kline 1977 {published data only}
  • Kline F, Burgoyne RW, Yamamoto J. Comparison of pimozide and trifluoperazine as once-daily therapy in chronic schizophrenic outpatients. Current Therapeutic Research 1977;21:768-78.
Kolivakis 1974 {published data only}
  • Kolivakis T, Azim H, Kingstone E. A double-blind comparison of pimozide and chlorpromazine in the maintenance care of chronic schizophrenic outpatients. Current Therapeutic Research 1974;16:998-1004.
Kudo 1972 {published data only}
  • Kudo Y. A double blind comparison of pimozide with carpipramine in schizophrenic patients. Acta Psychiatrica Belgica 1972;72:685-97.
  • Kudo Y, Fujiki A, Hino S, Kobayashi Y, Nagasaka G, Saitoh Y, et al. The double-blind study of pimozide for schizophrenia. Igaku No Ayumi 1972;82(11):726-37. [MEDLINE: 90144040; PMID 2694767]
McCreadie 1980 {published data only}
  • McCreadie R, Dingwall J, Wiles D, Heykants J. Intermittent pimozide versus fluphenazine decanoate as maintenance therapy in chronic schizophrenia. British Journal of Psychiatry 1980;137:510-7.
McCreadie 1982* {published data only}
McCreadie 1987 {published data only}
  • Scottish Schizophrenia Research Group. The Scottish First Episode Schizophrenia Study. VIII. Five-year follow-up: clinical and psychosocial findings. British Journal of Psychiatry 1992;161:496-500.
  • Scottish Schizophrenia Research Group: McCreadie RG, Wiles D, Grant S, Crocket GT, Mahmood Z, Livingston MG, et al. The Scottish First Episode Schizophrenia Study. VII. Two-year follow up. Acta Psychiatrica Scandinavica 1989;80:597-602.
  • The Scottish Schizophrenia Research Group. The Scottish First Episode Schizophrenia Study. II. Treatment: pimozide versus flupenthixol. British Journal of Psychiatry 1987;150:334-8.
  • The Scottish Schizophrenia Research Group. The Scottish First Episode Schizophrenia Study. III. Cognitive performance. British Journal of Psychiatry 1987;150:338-40.
  • The Scottish Schizophrenia Research Group. The Scottish First Episode Schizophrenia Study: one-year follow-up. British Journal of Psychiatry 1988;152:470-6.
McInnes 1978 {published data only}
  • McInnes EJ, Walker F, Snelling E. The observer interaction variable in evaluation of socialising properties of pimozide (Orap). New Zealand Medicine Journal 1978;87:170-2.
Nishikawa 1985 {published data only}
  • Nishikawa T, Tsuda A, Tanaka M, Koga I, Uchida Y. Prophylactic effects of neuroleptics in symptom-free schizophrenics: roles of dopaminergic and noradrenergic blockers. Biological Psychiatry 1985;20:1161-6.
Pecknold 1980 {published data only}
  • Pecknold JC, McClure DJ, Allan T, Wrzesinski L. Comparison of pimozide and chlorpromazine in acute schizophrenia. Canadian Journal of Psychiatry 1982;27:208-12.
Pinard 1972 {published data only}
  • Pinard G, Prenoveau Y, Fielsen W, Elie R, Bielman P, Lamontagne Y, et al. [Le pimozide et la reintegration sociale des schizophrenes chroniques]. Proceedings of the World Congress of Psychiatry, 28 November-4 December 1971; Ciudad de Mexico. 1971; Vol. 952.
  • Pinard G, Prenoveau Y, Fliesen W, Elie R, Bielmann P, Lamontagne Y, et al. Pimozide: a comparative study in the treatment of chronic schizophrenic patients. International Journal of Clinical Pharmacology 1972;6:22-7.
Silverstone 1984 {published data only}
  • Silverstone T, Cookson J, Ball R, et al. The relationship of dopamine receptor blockade to clinical response in schizophrenic patients treated with pimozide or haloperidol. Journal of Psychiatric Research 1984;18:255-68.
Vergara 1977 {published data only}
  • Vergara L, Amin MM, Ban TA. A standard (trifluoperazine) controlled clinical study with pimozide in the maintenance treatment of schizophrenic patients. Psychopharmacology Bulletin 1977;13:17-9.
Wilson 1982* {published data only}
  • Wilson LG, Roberts RW, Gerber CJ, Johnson MH. Pimozide versus chlorpromazine in chronic schizophrenia, a 52-week double-blind study of maintenance therapy. Journal of Clinical Psychiatry 1982;43:62-5.

References to studies excluded from this review

  1. Top of page
  2. AbstractRésumé scientifique
  3. Summary of findings
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' conclusions
  10. Acknowledgements
  11. Data and analyses
  12. Appendices
  13. What's new
  14. History
  15. Contributions of authors
  16. Declarations of interest
  17. Sources of support
  18. Differences between protocol and review
  19. Characteristics of studies
  20. References to studies included in this review
  21. References to studies excluded from this review
  22. References to studies awaiting assessment
  23. Additional references
  24. References to other published versions of this review
Baggio 1970 {published data only}
  • Baggio MA, Dias VS. Pimozide (R6238): clinical tests with a new incisive neuroleptic of long lasting action. Revista Brasileira Psiquiatrica 1970;4(4):189-93.
Baro 1972 {published data only}
  • Baro F, Van Lommel R, Dom R, De Mesmaecker L. Pimozide treatment of chronic schizophrenics as compared with haloperidol and penfluridol maintenance treatment-a multidisciplinary approach. Acta Psychiatrica Belgica 1972;72:199-214.
Bobon 1968 {published data only}
  • Bobon J, Collard J, Pinchard A, Goffioul L, Bobon DP, Devroye A. No English title available [Neuroleptiques a longue duree d'action II Etude de pilote du pimozide (R6238)]. Acta Neurologica Belgiana 1968;68:137-53.
  • DP Bobon, Devroye A, Goffioul L, Pinchard A. No English title available [Le pimozide: seize mois de follow up]. Acta Neurologica Belgiana 1968;68:888-94.
Brugmans 1968 {published data only}
  • Brugmans J. A multicentric clinical evaluation of pimozide: preliminary report. Acta Neurologica Psychiatrica Belgica 1968;68:875-87.
  • Janssen P, Brugmans J, Dony J, Schuermans V. An international double-blind clinical evaluation of pimozide. Journal Clinical Pharmacology and New Drugs 1972;12:26-34.
Cetin 2000 {published data only}
  • Cetin M, Ebrinc S. Risperidone versus pimozide for the treatment of monosymptomatic hypochondriacal psychosis. Journal of the European College of Neuropsychopharmacology 2000;10(3):s325.
Cheadle 1979 {published data only}
Chouinard 1979 {published data only}
  • Chouinard G, Annabel L, Lafontaine HL. Pimozide in the treatment of acute schizophrenia. Proceedings of the 132nd Annual Meeting of the American Psychiatric Association; 1979 May 12-18; Chicago, Illinois, USA. 1979.
Clark 1971 {published data only}
  • Clark M. Pimozide versus placebo. Psychopharmacology Bulletin 1971;7(1):35-7.
Crow 1986 {published data only}
  • Crow TJ, MacMillan JF, Johnson AL, Johnstone EC. II. A randomised controlled trial of prophylactic neuroleptic treatment. British Journal of Psychiatry 1986;148:120-7.
  • Johnstone EC, Owens DGC. Does early treatment have an effect on outcome?. Proceedings of the 10th European College of Neuropharmacology Congress; September 13-17, 1997; Vienna, Austria. 1997.
  • MacMillan JF, Crow TJ, Johnson AL, Johnstone EC. III. Short-term outcome in trial entrants and trial eligible patients. British Journal of Psychiatry 1986;148:128-33.
d'Elia 1974 {published data only}
  • d'Elia G, Perris C, Rapp W. Objective evaluation of EEG amplitude in chronic schizophrenic patients during a controlled trial of pimozide and trifluoperazine. Acta Psychiatrica Scandinavica Supplementum 1974;249:78-86.
Feinberg 1988 {published data only}
  • Feinberg SS, Stanley RK, Eliijovich LR, Fiszbein A, Opler LA. Pimozide treatment of the negative schizophrenic syndrome: an open trial. Journal of Clinical Psychiatry 1988;49(6):235-8.
Friedman 1997 {published data only}
Frussa Filho 1988 {published data only}
  • Frussa Filho R, Palermo Neto J. No English title available [Sobre os efeitos colaterais extrapyramidais dos neurolepticos]. Farmacologia Clinica 1988;96(6):389-92.
Garton 1979 {published data only}
Gomez Perez 1994 {published data only}
  • Gomez Perez JC. No English title available [Dismorfofobia: Pasado y presente de un trastorno centenario]. Actas Luso-Espanol Neurologia Psiquiatria 1994;22(2):83-8.
Hass 1982 {published data only}
  • Haas S, Beckmann H. Pimozide and haloperidol in a double-blind study in acutely schizophrenic patients. Arzneimittel Forschung 1982;32(8):888.
Holl 1992 {published data only}
Huber 1983 {published data only}
Ibarra 1996 {published data only}
  • Ibarra HS, Concha SJ, Coronel AR, Cruz PR, Alarcon NA. Pimozide in the treatment of delusional disorder. Proceedings of the 10th World Congress of Psychiatry; 1996 Aug 23-28; Madrid, Spain. 1996.
Johnstone 1997 {published data only}
  • Johnstone EC, Crow TJ, Frith CD, Owens DG. The Northwick park "functional" psychosis study: diagnosis and treatment response. Lancet 1988;2:119-25.
Kenway 1971 {published data only}
  • Kenway AK, Masheter HC. Pimozide compared with fluphenazine in schizophrenia. British Journal of Clinical Practice 1971;25:69-72.
Kenway 1973 {published data only}
  • Kenway AK. A double-blind comparison of pimozide and haloperidol in the treatment of recurrent anxiety states. British Journal of Clinical Practice 1973;27:67-8.
Konig 1971 {published data only}
  • Konig L, Lange E, Mucha H, Winkler J, Kunath B. No English title available [Klinische Moglichkeiten der Therapiebeurteilung in der Pharmakotherapie am Beissspel der Wirksamkeitsprufung eines neuen Langzeit Neuroleptikums pimozide]. Psychiatry Neurology Medicine and Psychology 1971;23:359-67.
Koo 1996 {published data only}
Krumholz 1970 {published data only}
  • Krumholz. Pimozide versus standards versus placebo. Psychopharmacology Bulletin 1970;7(2):68-70.
Lapierre 1976 {published data only}
  • Lapierre YD, Lavallee J. A controlled pimozide, fluphenazine and group psychotherapy study of chronic schizophrenics. Psychiatric Journal of the University of Ottawa 1976;1:8-13.
  • Lapierre YD, Lavallee J. Pimozide and the social behavior of schizophrenics. Current Therapeutic Research 1975;18(1):181-8.
Lehmann 1970 {published data only}
  • Lehmann B. Pimozide versus fluphenazine. Psychopharmacology Bulletin 1970;7(2):61-3.
Mahal 1975 {published data only}
  • Mahal AS, Janakiramaiah N. A double blind placebo controlled trial of pimozide (R6238) on 49 hospitalized chronic schizophrenics. Indian Journal of Psychiatry 1975;17(1):45-55.
Marshall 1971 {published data only}
  • Marshall WK. Pimozide (Orap): a multicentre study. Clinical Trials Journal 1971;7(Suppl 2):49-54.
Masiak 1976 {published data only}
  • Masiak M, Majczak A, Hascewicz-Rzecka M. Pimozide and piportil in the treatment of chronic schizophrenia (preliminary communication) [Badania nad zastosowaniem pimozidu i piportilu w leczeniu chorych na przewlekla schizofrenie]. Psychiatrica Poland 1976;10(6):655-60.
McCoy 1992 {published data only}
  • McCoy LM, Schwazkopf SB, Martin D. Rapid response to pimozide in treatment resistant delusional disorder. Annals of Clinical Psychiatry 1992;4(2):95-8.
McCreadie 1978 {published data only}
McCreadie 1983 {published data only}
  • McCreadie RG, McKane JP, Mackie M. Weekly pimozide versus fluphenazine decanoate in schizophrenic out- and day-patients. British Journal of Psychiatry 1983;143:97-8.
Meltzer 1986 {published data only}
Moller 1994 {published data only}
  • Moller HJ, Van Praag HM, Aufdembrinke B, Bailey P, Barnes TRE, Beck J, et al. Negative symptoms in schizophrenia: considerations for clinical trials: working group on negative symptoms in schizophrenia. Psychopharmacology 1994;115:221-8.
Morris 1970 {published data only}
Neziroglu 1997 {published data only}
Opler 1994 {published data only}
Opler 1995 {published data only}
Phillips 1996 {published data only}
Pinals 1996 {published data only}
Poldinger 1976 {published data only}
  • Poldinger W. Application of the neuroleptic pimozide (ORAP-R6238) for tranquillizer indications in a controlled study. International Pharmacopsychiatry 1976;11:16-24.
Reyntjens 1972 {published data only}
  • Reyntjens AM, Van Mierlo FP. A comative double-blind trial of pimozide in stress-induced psychic and functional disorders. Current Medical Research Opinion 1972;1:116-22.
Ruiz 1975 {published data only}
  • Ruiz Ruiz M, Miro Quintana L, Sentis Vilalta J. A clinical study with pimozide in chronic schizophrenics [Estudio clinico con pimocide en las esquizophrenias de evolucion cronica]. Revista de Psiquiatria y Psicologia Medica de Europa y America Latinas 1975;12(4):247-56. [PsycINFO 65-08563]
Sims 1975 {published data only}
  • Sims AC, Burnside IG. Activity in chronic schizophrenic patients: comparison of pimozide with fluphenazine in a double blind trial. Psychological Medicine 1975;5:161-4.
Singh 1971 {published data only}
  • Singh AN. Evaluation of clinical efficacy of pimozide as maintenance therapy in chronic schizophrenic patients. Current Therapeutic Research 1971;13(11):695-705.
Smythies 1974 {published data only}
Srinivasan 1994 {published data only}
  • Srinivasan TN, Sureshi TR, Jayaram V, Fernandez MP. Nature and treatment of delusional parasitosis: a different experience in India. International Journal of Dermatology 1994;33(12):851-5.
Sterkmans 1968 {published data only}
  • Sterkmans P, Brugmans J, Gevers F. The clinical efficacy of pimozide in chronic psychotic patients. Clinical Trials Journal 1968;5(4):1107-12.
Sugerman 1971 {published data only}
Svestka 1972 {published data only}
Tegeler 1983 {published data only}
Tueth 1983 {published data only}
  • Teuth MJ, Cheong JA. Clinical uses of pimozide. Southern Medical Journal 1993;86(3):344-9.
Ungvari 1986 {published data only}
Vanelle 1996 {published data only}
  • Vanelle JM. Profile of the action of neuroleptics in deficit schizophrenia [Profil d'action des neuroleptiques dans les schizophrenies defictaires]. L'Encephale 1996;22(Suppl 2):33-9.
van Kammen 1982 {published data only}
  • van Kammen DP, Docherty JP, Marder SR, Schulz SC, Dalton L, Bunney WE Jr. Antipsychotic effects of pimozide in schizophrenia: treatment response prediction with acute dextroamphetamine response. Archives of General Psychiatry 1982;39:261-6.
van Kammen 1987 {published data only}
  • Van Kammen DP, Hommer DW, Malas KL. Effect of pimozide on positive and negative symptoms in schizophrenic patients: are negative symptoms state dependent?. Neuropsychobiology 1987;18:113-7.
Van Wyck 1972 {published data only}
  • Van Wyk AJ, Grove JJ. Pimozide: an effective, once daily oral therapy for schizophrenia. South African Medical Journal 1972;46:515-7.
Watt 1983 {published data only}
  • Watt DC, Katz K, Shepherd M. The natural history of schizophrenia: a 5-year prospective follow-up of a representative sample of schizophrenics by means of a standardized clinical and social assessment. Psychological Medicine 1983;13:663-70.
Welbel 1979 {published data only}
  • Welbel L. Differences in the clinical effects of various neuroleptics. Psychiatrica Poland 1980;XIV(2):113-8.

Additional references

  1. Top of page
  2. AbstractRésumé scientifique
  3. Summary of findings
  4. Background
  5. Objectives
  6. Methods
  7. Results
  8. Discussion
  9. Authors' conclusions
  10. Acknowledgements
  11. Data and analyses
  12. Appendices
  13. What's new
  14. History
  15. Contributions of authors
  16. Declarations of interest
  17. Sources of support
  18. Differences between protocol and review
  19. Characteristics of studies
  20. References to studies included in this review
  21. References to studies excluded from this review
  22. References to studies awaiting assessment
  23. Additional references
  24. References to other published versions of this review
Adams 2007
Altman 1996
  • Altman DG, Bland JM. Detecting skewness from summary information. BMJ 1996;313:1200. [PIM020600]
Andreasen 1982
  • Andreasen NC. Negative symptoms in schizophrenia: definition and reliability. Archives of General Psychiatry 1982;39(7):784-8.
Begg 1996
  • Begg C, Cho M, Eastwood S, Horton R, Moher D, Olkin I, et al. Improving the quality of randomized controlled trials: the CONSORT statement. JAMA 1996;276:637-9.
Bland 1997
  • Bland JM. Statistics notes: trials randomised in clusters. BMJ 1997;315:600.
BNF
  • BNF. http://www.bnf.org/bnf/index.htm.
BNF 2012
  • Royal Pharmaceutical Society of Great Britain. British National Formulary. Vol. 63, The Pharmaceutical Press, 2012.
Boissel 1999
  • Boissel JP, Cucherat M, Li W, Chatellier G, Gueyffier F, Buyse M, et al. The problem of therapeutic efficacy indices. 3. Comparison of the indices and their use [Apercu sur la problematique des indices d'efficacite therapeutique, 3: comparaison des indices et utilisation. Groupe d'Etude des Indices D'efficacite]. Therapie 1999;54(4):405-11. [PUBMED: 10667106]
Caldwell 1992
Chouinard 1980
  • Chouinard G, Ross-Chouinard A, Annable L, Jones B. The Extrapyramidal Symptom Rating Scale. Canadian Journal of Neurological Sciences 1980;7:233.
Cotton 2006
  • Cotton S, Yuen HP, Berger G, McGorry P. Last observation carried forward in psychiatric research: a flawed gold standard?. Schizophrenia Research 2006;81(Suppl):227.
CSM 1990
  • Committee on Safety of Medicines. Cardiotoxic effects of pimozide. Current Problems 1990;29:1-2.
Davies 2007
  • Davies EJ. Developmental aspects of schizophrenia and related disorders: possible implications for treatment strategies. Advances in Psychiatric Treatment 2007;13:384-91.
De Hert 2011
  • De Hert M, Correll CU, Bobes J, Cetkovich-Bakmas M, Cohen D, Asai I, et al. Physical illness in patients with severe mental disorders. I. Prevalence, impact of medications and disparities in health care. World Psychiatry 2011;10:52-77.
Deeks 2000
  • Deeks J. Issues in the selection for meta-analyses of binary data. Proceedings of the 8th International Cochrane Colloquium; 2000 October 25-28; Cape Town. Cape Town: The Cochrane Collaboration, 2000.
Divine 1992
Donner 2002
Egger 1997
  • Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629-34.
Elbourne 2002
  • Elbourne D, Altman DG, Higgins JPT, Curtina F, Worthingtond HV, Vaile A. Meta-analyses involving cross-over trials: methodological issues. International Journal of Epidemiology 2002;31(1):140-9.
Feighner 1972
  • Feighner JP, Robins E, Guze SB, Woodruff RA, Winokur G, Munos R. Diagnostic criteria for the use in psychiatric research. Archives of General Psychiatry 1972;26:57-63.
Furukawa 2006
  • Furukawa TA, Barbui C, Cipriani A, Brambilla P, Watanabe N. Imputing missing standard deviations in meta-analyses can provide accurate results. Journal of Clinical Epidemiology 2006;59(7):7-10.
Gulliford 1999
  • Gulliford MC. Components of variance and intraclass correlations for the design of community-based surveys and intervention studies: data from the Health Survey for England 1994. American Journal of Epidemiology 1999;149:876-83.
Guy 1976
  • Guy W. Early Clinical Drug Evaluation (ECDEU) Assessment Manual for Psychopharmacology. Washington, DC: National Institute of Mental Health, Publication No.76-338, 1976.
Higgins 2003
Higgins 2005
  • Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Interventions 4.2.5 [updated May 2005]. Cochrane Library. Vol. 3, Chichester, UK: John Wiley & Sons, Ltd, 2005.
Higgins 2011
  • Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.2 [updated September 2011]. The Cochrane Collaboration, 2011. www.cochrane-handbook.org..
ICD 10 1994
  • World Health Organisation. ICD-10 Classification of Mental and Behavioral Disorders. London: Churchill Livingstone, 1991.
Jadad 1996
  • Jadad A, Moore A, Carrol D, Jenkinson C, Reynolds DJM, Gavanagh DJ, et al. Assessing the quality of reports of randomised controlled trials: is blinding necessary?. Controlled Clinical Trials 1996;17:1-12.
Janssen 1998
  • Janssen PA. Function and dysfunction of the basal ganglia. In: Bain TA, Healy D, Shorter E editor(s). The Rise of Psychopharmacology and the Story of CINP. Budapest: Animula Publishing House, 1998.
Jusic 1994
Kaplan 1994
  • Kaplan HI, Sadock BJ, Grebb JA. Other psychotic disorders. In: Kaplan HI, Sadock BJ, Grebb JA editor(s). Synopsis of Psychiatry. London: Williams and Wilkins, 1994:487-512.
Katz 1963
  • Katz MM, Lyerly SB. Methods for measuring adjustment and social behaviour in the community. I. Rationale, description, discriminative validity and scale development. Psychological Reports 1963;13:503-35.
Kay 1986
  • Kay SR, Opler LA, Fiszbein A. Positive and Negative Syndrome Scale (PANSS) Manual. North Tonawanda, NY: Multi-Health Systems, 1986.
Kay 1987
  • Kay SR, Fiszbein A, Opler LA. The Positive and Negative Syndrome Scale (PANSS) for schizophrenia. Schizophrenia Bulletin 1987;13(2):261-76.
King 1995
  • King DJ. Neuroleptics and schizophrenia. Seminars in Clinical Psychopharmacology 1995:259-327.
Leucht 2005
Leucht 2005a
  • Leucht S, Kane JM, Kissling W, Hamann J, Etschel E, Engel R. Clinical implications of Brief Psychiatric Rating Scale scores. British Journal of Psychiatry 2005;187:366-71. [PUBMED: 16199797]
Leucht 2007
Marshall 2000
  • Marshall M, Lockwood A, Bradley C, Adams C, Joy C, Fenton M. Unpublished rating scales: a major source of bias in randomised controlled trials of treatments for schizophrenia. British Journal of Psychiatry 2000;176:249-52.
Moher 1998
  • Moher D, Pham B, Jones A, Cook DJ, Jadad AR, Moher M, et al. Does quality of reports of randomised trials affect estimates of intervention efficacy reported in meta-analyses?. Lancet 1998;352:609-13.
Moher 2001
Montgomery 1979
Munro 1980
Opler 1991
Overall 1962
  • Overall JE, Gorham DR. The Brief Psychiatric Rating Scale. Psychological Reports 1962;10:799-812.
Platt 1980
  • Platt S, Weyman A, Hirsch S, Hewett S. The Social Behaviour Assessment Schedule (SBAS): rationale, contents, scoring and reliability of a new interview schedule. Social Psychiatry 1980;15:43-55.
Quraishi 1999
Radhakrishnan 2012
  • Radhakrishnan R, Butler R, Head L. Dementia in schizophrenia. Advances in Psychiatric Treatment 2012;18:144-53.
Schneider 1983
  • Schneider LC, Struening EL. SLOF: a behavioral rating scale for assessing the mentally ill. Social Work Research & Abstracts 1983;19(3):9-21.
Schulz 1994
  • Schulz KF, Chalmers I, Grimes DA, Altman DG. Assessing the quality of randomization from reports of controlled trials published in obstetrics and gynecology journals. JAMA 1994;272:125-8.
Schulz 1995
  • Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias: dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 1995 1995;273:408-12.
Schünemann 2008
  • Schünemann HJ, Oxman AD, Vist GE, Higgins JPT, Deeks JJ, Glasziou P, et al. Chapter 12: Interpreting results and drawing conclusions. In: Higgins JPT, Green S editor(s). Cochrane Handbook for Systematic Reviews of Interventions. Copenhagen: The Cochrane Collaboration, 2008:359-83.
Silverstone 1995
  • Silverstone T, Turner P. Drug Treatment in Psychiatry. 5th Edition. London: Routledge, 1995.
Simpson 1970
  • Simpson GM, Angus JWS. A rating scale for extrapyramidal side effects. Acta Psychiatria Scandinavica Supplementum 1970;212:11-9.
Turner 1990
  • Turner WM, Tsuang MT. The impact of substance abuse on the course and outcome of schizophrenia. Schizophrenia Bulletin 1990;16:18.
Udabe 1998
  • Udabe RU. Psychopharmacology in Argentina. In: Bain TA, Healy D, Shorter E editor(s). The Rise of Psychopharmacology and the Story of CINP. Budapest: Animula Publishing House, 1998.
Ukoumunne 1999
  • Ukoumunne OC, Gulliford MC, Chinn S, Sterne JAC, Burney PGJ. Methods for evaluating area-wide and organistation-based intervention in health and health care: a systematic review. Health Technology Assessment 1999;3(5):1-75.
Wing 1974
  • Wing JK, Cooper J, Sartorious N. The Measurement and Classification of Symptoms. Cambridge: Cambridge University Press, 1974.
Winokur 1977
Xia 2009
  • Xia J, Adams CE, Bhagat N, Bhagat V, Bhoopathi P, El-Sayeh H, et al. Loss to outcomes stakeholder survey: the LOSS study. Psychiatric Bulletin 2009;33(7):254-7.
Yaryura-Tobias 1998
  • Yaryura-Tobias JA. Serotonin and obsessive compulsive disorder. In: Bain TA, Healy D, Shorter E editor(s). The Rise of Psychopharmacology and the Story of CINP. Budapest: Animula Hublishing House, 1998.