Anticholinergic Discontinuation for Antipsychotic-Induced Extra-Pyramidal Symptoms

Anticholinergic discontinuation for antipsychotic- induced extra-pyramidal symptoms

Item Type Article

Authors Albert, Michael

Citation Pereira, S. R. & Albert, M. (2017). Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms. Cochrane Database of Systematic Reviews, -(3), pp.14.

Download date 25/09/2021 03:40:14

Link to Item http://hdl.handle.net/20.500.12904/9804 Cochrane Database of Systematic Reviews

Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms (Protocol)

Pereira SR, Albert M

Pereira SR, Albert M. Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms. Cochrane Database of Systematic Reviews 2017, Issue 3. Art. No.: CD012525. DOI: 10.1002/14651858.CD012525. www.cochranelibrary.com

Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms (Protocol) Copyright © 2017 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. TABLE OF CONTENTS HEADER...... 1 ABSTRACT ...... 1 BACKGROUND ...... 1 OBJECTIVES ...... 2 METHODS ...... 3 ACKNOWLEDGEMENTS ...... 9 REFERENCES ...... 10 CONTRIBUTIONSOFAUTHORS ...... 11 DECLARATIONSOFINTEREST ...... 11 SOURCESOFSUPPORT ...... 12

Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms (Protocol) i Copyright © 2017 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. [Intervention Protocol] Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms

Simi R Pereira1, Michael Albert2

1General Adult Psychiatry, Norfolk and Suffolk NHS Foundation Trust, Norwich, UK. 2Department of Psychiatry, Highbury Hospital, Nottingham, UK

Contact address: Simi R Pereira, General Adult Psychiatry, Norfolk and Suffolk NHS Foundation Trust, Drayton High Road, Norwich, Norfolk, NR5 6BE, UK. [email protected], [email protected].

Editorial group: Cochrane Schizophrenia Group. Publication status and date: New, published in Issue 3, 2017.

Citation: Pereira SR, Albert M. Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms. Cochrane Database of Systematic Reviews 2017, Issue 3. Art. No.: CD012525. DOI: 10.1002/14651858.CD012525.

ABSTRACT

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

To assess the effects of discontinuation compared with continued use of anticholinergic drugs prescribed for antipsychotic-induced extrapyramidal symptoms in people with schizophrenia and related disorders including schizophreniform disorder, delusional disorder, psychosis and schizoaffective disorder.

BACKGROUND ing antipsychotics. Parkinsonism is a group of symptoms, usually emerging within a few days after starting antipsychotic treatment or a dose increase, which resemble idiopathic Parkinson’s disease, including tremor, bradykinesia (slow movement), rigidity (limb Description of the condition stiffness), mask-like face, hypersalivation and postural instability. People with severe mental illness are often prescribed antipsy- The chronic EPSEs such as tardive dyskinesia are more commonly chotics to treat their symptoms. Such medications exert their ther- seen after years of antidopaminergic (e.g. antipsychotic) therapy apeutic effects by blocking the dopamine receptors. However, and the pathophysiology is uncertain. Tardive dyskinesia symp- one of the side effects of blocking the dopamine receptors is the toms include restlessness and involuntary movement of groups of emergence of extrapyramidal side effects (EPSEs) due to the anti- muscles in the tongue, lips, face, trunks and extremities. Unlike dopaminergic effects of antipsychotics on the basal ganglia. EPSEs the acute EPSEs, it may not be resolved by withdrawing the treat- can be acute and chronic. Acute EPSEs are dystonias, parkinson- ment. ism and akathisia, and those classified as chronic include tardive Although most antipsychotics can induce EPSEs, the prevalence dyskinesia and tardive dystonia (American Psychiatric Association and incidence varies between antipsychotics. It is difficult to state 2006). an exact figure for prevalence of EPSEs, the cumulative incidence Dystonia is an acute involuntary muscle contraction or spasm and ranges from 7.7% (olanzapine) to 32.8% (depot typical drugs) akathisia is a feeling of inner restlessness and the urge to move. (Novick 2010). The incidence of tardive dyskinesia ranges from They are both often seen within a few hours to a few weeks of start-

Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms (Protocol) 1 Copyright © 2017 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. 2.8% (olanzapine) to 11.1% (depot typical drugs) (Novick 2010). How the intervention might work Both typical and atypical antipsychotics are included in the World Dopamine normally suppresses acetylcholine activity; therefore Health Organization (WHO) Model List of Essential Medicines removal of dopamine inhibition causes an increase in release (WHO 2015); however, people living in low and middle income of acetylcholine and therefore cholinergic activity (Stahl 2008). countries have a limited choice of medication, which necessitates Acetylcholine is an excitatory neurotransmitter which can cause the use of older, cheaper drugs such as haloperidol, chlorpromazine EPSEs if it is in excess. This is a possible explanation of how and fluphenazine, with a higher potency for EPSE. antipsychotics, whose main action is dopamine blockade, cause People with parkinsonism report slowness of thinking or ‘feeling EPSEs: by removing the dopamine inhibition and thereby in- like a zombie’, while those with akathisia describe inner restlessness creased cholinergic activity and subsequent EPSEs. Anticholiner- and unease. While individuals are often unaware of the movements gic drugs block the acetylcholine receptors thereby overcoming the caused by tardive dyskinesia and are usually not distressed by them, excessive cholinergic activity. Discontinuation of anticholinergics relatives often find them distressing. These movements can be very may cause re-emergence of EPSEs as the acetylcholine receptors obvious to onlookers and can set people apart socially, possibly cease to be blocked, resulting in excessive cholinergic activity. adding to the stigma of severe psychiatric illness. Extrapyramidal symptoms result in social and functional disability; the suffering and stigmatisation associated with them is likely to remain with Why it is important to do this review us for a long time to come (Gervin 2000). There is no current evidence from systematic reviews to support prescribing anticholinergic drugs in the absence of extrapyramidal side effects, and they have various undesirable side effects. This review aims to assess whether anticholinergic medications are required for long-term treatment of extrapyramidal symptoms Description of the intervention by reviewing the effect of discontinuation. There are no previous Cochrane reviews comparing the discontinuation of anticholin- Anticholinergic (antimuscarinic) drugs are prescribed alongside ergic medication to continued use. The clinical relevance of this antipsychotic drugs to help prevent unwanted extrapyramidal side review is to assess if we are making people worse by possible emer- effects often associated with antipsychotics. Anticholinergics li- gence or re-emergence of EPSE(s) by stopping adjunct anticholin- censed for use in the United Kingdom are orphenadrine, procycli- ergic medication, or if adjunct anticholinergic medication needs dine and trihexyphenidyl (BNF 2015). Biperiden is the anticholin- to be continued long term. geric drug on the World Health Organization (WHO) Model List There are several Cochrane reviews on anticholinergic med- of Essential Medicines (WHO 2015). There are other anticholin- ication for extrapyramidal side effects: these include ’Anti- ergic medications available, for example benztropine, dexetemide, cholinergics versus placebo for neuroleptic-induced parkinson- etybenzatropine and profenamine (also known as ethopropazine). ism’ (Dickenson 2014a); ’Anticholinergics (various) for neurolep- Anticholinergic drugs can aggravate the anticholinergic side ef- tic-induced parkinsonism’ (Dickenson 2014b); ’Anticholinergics fects of antipsychotic drugs which can precipitate glaucoma, dry for neuroleptic-induced akathisia’ (Rathbone 2006); and ’Anti- mouth or urinary retention. They can also cause constipation and cholinergic medication for neuroleptic-induced tardive dyskine- drowsiness and have been associated with an increase in ’positive’ sia’ (Soares-Weiser 1997). A recommendation which stems from psychotic symptoms (Johnstone 1983), cognitive decline (Bottiggi ’Anticholinergic versus placebo for neuroleptic-induced parkin- 2006), and with worsening of tardive dyskinesia (Yassa 1988). An- sonism’ is to review the effect of discontinuation of anticholinergic ticholinergic medications can be abused for their stimulating and medication, which has not previously been studied in a Cochrane euphoriant effects (Pullen 1984), with one study showing that systematic review. The reason for this is that many studies cur- 33% of people prescribed anticholinergic medication had abused rently being excluded from the parkinsonism reviews (Dickenson them within the previous month (Buhrich 2000). 2014a and Dickenson 2014b) are not specific as to the type of Guidelines recommend reducing or switching antipsychotics or EPSE that the participants were suffering from at the start of the adding short-term anticholinergic medications as treatment for study. The discontinuation of anticholinergic drugs may also re- those presenting with antipsychotic-induced parkinsonism (Taylor sult in the emergence of other EPSEs. 2012). Dystonia can be treated with anticholinergic drugs given orally, intramuscularly or intravenously (Taylor 2012). Anticholinergic drugs should be reduced or discontinued if the patient is reporting intolerable side effects as described above, as OBJECTIVES with any medication-induced side effect. Discontinuation of anticholinergic drugs prescribed along with antipsychotic drugs may To assess the effects of discontinuation compared with contin- or may not precipitate extrapyramidal symptoms. ued use of anticholinergic drugs prescribed for antipsychotic-in-

Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms (Protocol) 2 Copyright © 2017 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. duced extrapyramidal symptoms in people with schizophrenia and Types of outcome measures related disorders including schizophreniform disorder, delusional We aim to divide all outcomes into immediate-term (less than 24 disorder, psychosis and schizoaffective disorder. hours), short-term (one day to one week) and medium-term (over one week).

METHODS Primary outcomes

Criteria for considering studies for this review

1. EPSE symptoms

Types of studies 1.1 Clinically important change in EPSE (parkinsonism, dystonia, acute akathisia) (as defined by individual studies) - short term We will consider all relevant randomised controlled trials. If a trial is described as ’double blind’ but implies randomisation, we will include such trials in a sensitivity analysis (see Sensitivity analysis). We will exclude quasi-randomised studies, such as those allocating 2. General functioning by alternate days of the week. Where people are given additional 2.1 Clinically important change in general functioning (as defined treatments along with discontinuation of anticholinergics, we will only include data if the adjunct treatment is evenly distributed be- by individual studies) long term tween groups and it is only the discontinuation of anticholinergics that is randomised. 3. Quality of life 3.1 Clinically important change in quality of life (as defined by Types of participants People with schizophrenia or related disorders, including individual studies) long term schizophreniform disorder, schizoaffective disorder and delusional disorder who have been prescribed anticholinergic drugs along with antipsychotic drugs, who are currently either experiencing or Secondary outcomes not experiencing EPSE(s).

Types of interventions 1. EPSE symptoms 1.1 Clinically important change in EPSE (dystonia, acute and chronic akathisia, parkinsonism) (as deﬁned by individual studies) 1. Discontinuation of anticholinergic or substitution with a placebo to aid discontinuation medium to long term 1.2 Any change in EPSE (parkinsonism, tardive dyskinesia, Discontinuation of anticholinergic drug prescribed along with continued antipsychotic or substitution by a placebo, active or in- akathisia) (as deﬁned by individual studies) short, medium or active. Placebo could be any dose, frequency or route of adminis- long term tration taken with any type/dose/route/frequency of antipsychotic.

2. General functioning 2. Continuation of an anticholinergic Any anticholinergic drug, any dose, any frequency, any route of 2.1 Any change in general functioning (as deﬁned by individual administration, taken with any type/dose/route/frequency of an- studies) tipsychotic. 2.1 Average endpoint/change scores on general functioning scale

3. Different methods of discontinuation of anticholinergics 3. Quality of life Various methods of how the anticholinergic is discontinued (sud- 3.1 Any change in quality of life (as deﬁned by individual studies) den, gradual etc.). 3.2 Average endpoint/change scores on general quality-of-life scale

Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms (Protocol) 3 Copyright © 2017 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. 4. Mental state 4. Mental state: clinically important change in general mental 4.1 Clinically important change in general mental state (as defined state(as defined by individual studies) - long term by individual studies) 5. Adverse effects: clinically important adverse effect - long 4.2 Any change in general mental state (as defined by individual term. studies) 6. Hospitalisation: number of participants requiring 4.3 Average endpoint/change scores on general mental state scale hospitalisation. 4.4 Clinically important change in specific symptoms (as defined 7. Leaving the study early: any reason by individual studies) 4.5 Any change in specific symptoms (as defined by individual studies) 4.6 Average endpoint/change scores on specific symptom scales Search methods for identification of studies

5. Adverse effects/event 5.1 Number of participants with at least one adverse effect Electronic searches 5.2 Clinically important speciﬁc adverse effects (e.g. aggravation of the anticholinergic side effects of the antipsychotic drugs, cognitive decline, worsening of tardive dyskinesia, cardiac effects, movement disorders, worsening of psychosis, sedation, seizures, weight gain, Cochrane Schizophrenia Group’s Trials Register effects on white blood cell count, anticholinergic misuse) The Trials Search Co-ordinator (TSC) will search Cochrane 5.3 Average endpoint/change score on adverse effects scale Schizophrenia’s study-based Register of Trials using the following 5.4 Death (suicide or other causes) search strategy: *Anticholinergic Medication* in Intervention Field of STUDY In such a study-based register, searching the major concept re- 6. Hospitalisation trieves all the synonym keywords and relevant studies because all 6.1 Number of participants requiring hospitalisation the studies have already been organised based on their interven- 6.2 Length of stay in hospital tions and linked to the relevant topics. Cochrane Schizophrenia’s Register of Trials is compiled by systematic searches of major resources (including AMED, BIOSIS, 7. Leaving the study early CINAHL, Embase, MEDLINE, PsycINFO, PubMed, and reg- 7.1 Any reason istries of clinical trials) and their monthly updates, handsearches, 7.2 Due to adverse effect/event grey literature, and conference proceedings (see Group’s Module). 7.3 Due to inefﬁcacy of treatment There is no language, date, document type, or publication status’ limitations for inclusion of records into the register.

’Summary of findings’ table We will use the GRADE approach to interpret findings ( Searching other resources Schünemann 2008); and we will use GRADEPRO to import data from Review Manager to create ’Summary of findings’ tables. These tables provide outcome-specific information concern- ing the overall quality of evidence from each included study in the 1. Reference searching comparison, the magnitude of effect of the interventions exam- ined, and the sum of available data on all outcomes we rated as We will inspect references of all included studies for further rele- important to patient care and decision making. We aim to select vant studies. the following main outcomes for inclusion in the ’Summary of findings’ table: 1. EPSE symptoms: clinically important change in EPSE (as 2. Personal contact defined by individual studies) - short term 2. General functioning: clinically important change in general We will contact the first author of each included study for infor- functioning (as defined by individual studies) - long term. mation regarding unpublished trials. We will note the outcome of 3. Quality of life: clinically important change in quality of life this contact in the ’Included studies’ or ’Studies awaiting classifi- (as defined by individual studies) - long term cation’ sections.

Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms (Protocol) 4 Copyright © 2017 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Data collection and analysis or shown to be reliable. However there are exceptions: we will include sub-scores from mental state scales measuring positive and negative symptoms of schizophrenia. Selection of studies Ideally the measuring instrument should either be i. a self-report SP will independently inspect citations from the searches and iden- or ii. completed by an independent rater or relative (not the thera- tify relevant abstracts. MA will independently re-inspect a random pist). We realise that this is not often reported clearly: in ’Descrip- 20% sample to ensure reliability. Where disputes arise, we will tion of studies’ we will note if this is the case or not. acquire the full report for more detailed scrutiny. SP will obtain and independently inspect full reports of the abstracts meeting the review criteria. Again, MA will independently re-inspect a random 20% of the full reports in order to ensure reliable selection. Where 2.3 Endpoint versus change data it is not possible to resolve disagreement by discussion, we will at- There are advantages of both endpoint and change data. Change tempt to contact the authors of the study for clarification. We will data can remove a component of between-person variability from include studies that meet the review inclusion criteria and report the analysis. On the other hand calculation of change needs two as- useable data. We will enter studies that meet the review inclusion sessments (baseline and endpoint) which can be difficult in unsta- criteria but do not present useable data as excluded studies. ble and difficult-to-measure conditions such as schizophrenia. We have decided to primarily use endpoint data, and only use change data if the former are not available. We will combine endpoint Data extraction and management and change data in the analysis as we prefer to use mean differences (MD) rather than standardised mean differences throughout (Higgins 2011). 1. Extraction Review authors SP will extract data from all included studies. In addition, to ensure reliability, MA will independently extract data from a random sample of these studies, comprising 10% of the 2.4 Skewed data total. We will discuss any disagreement and document decisions. Continuous data on clinical and social outcomes are often not If necessary, we will contact authors of studies for clarification. We normally distributed. To avoid the pitfall of applying parametric will extract data presented only in graphs and figures whenever tests to non-parametric data, we aim to apply the standards below possible, but will only include if two review authors independently to all data before inclusion. have the same result. We will attempt to contact authors through Please note: we will enter data from studies of at least 200 par- an open-ended request in order to obtain missing information ticipants, for example, in the analysis irrespective of the following or for clarification whenever necessary. If studies are multi-centre rules, because skewed data pose less of a problem in large studies. we will, where possible, extract data relevant to each component We will also enter change data as when continuous data are pre- centre separately. sented on a scale that includes a possibility of negative values (such as change data), it is difficult to tell whether data are skewed or not. We will present and enter change data into statistical analyses 2. Management For endpoint data: (a) when a scale starts from the nite number zero, we will subtract the lowest possible value from the mean, and divide this by the 2.1 Forms standard deviation. If this value is lower than 1, it strongly suggests We will extract data onto standard, simple forms. a skew and we will exclude such data. If this ratio is higher than 1 but below 2, there is suggestion of skew. We will enter these data and test whether their inclusion or exclusion would change 2.2 Scale-derived data the results substantially. Finally, if the ratio is larger than 2 we We will include continuous data from rating scales only if: will include these data because skew is less likely (Altman 1996; Higgins 2011). a) the psychometric properties of the measuring instrument have b) if a scale starts from a positive value (such as the Positive and been described in a peer-reviewed journal (Marshall 2000); and Negative Syndrome Scale (PANSS), which can have values from b) the measuring instrument has not been written or modified by 30 to 210) (Kay 1986), we will modify the calculation described one of the trialists for that particular trial. above to take the scale starting point into account. In these cases c) the instrument should be a global assessment of an area of func- skew is present if 2 SD > (S S min), where S is the mean score − tioning and not sub-scores which are not, in themselves, validated and ’S min’ is the minimum score.

Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms (Protocol) 5 Copyright © 2017 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. 2.5 Common measure For binary outcomes we will calculate a standard estimation of the To facilitate comparison between trials, we intend to convert vari- risk ratio (RR) and its 95% confidence interval (CI). It has been ables that can be reported in different metrics, such as days in hos- shown that RR is more intuitive than odds ratios (Boissel 1999); pital (mean days per year, per week or per month) to a common and that odds ratios tend to be interpreted as RR by clinicians metric (e.g. mean days per month). (Deeks 2000). The statistics of number needed to treat for an additional beneficial outcome (NNTB) and number needed to treat for an additional harmful outcome (NNTH) with their confidence 2.6 Conversion of continuous to binary intervals are intuitively attractive to clinicians but are problematic Where possible, we will make efforts to convert outcome measures both in accurate calculation in meta-analyses and interpretation to dichotomous data. This can be done by identifying cut-off (Hutton 2009). For binary data presented in the ’Summary of points on rating scales and dividing participants accordingly into findings’ table/s, where possible we will calculate illustrative com- ’clinically improved’ or ’not clinically improved’. It is generally parative risks. assumed that if there is a 50% reduction in a scale-derived score such as the Brief Psychiatric Rating Scale (BPRS) (Overall 1962), or the Positive and Negative Syndrome Scale (PANSS) (Kay 1986); 2. Continuous data this could be considered as a clinically significant response (Leucht For continuous outcomes we will estimate mean difference (MD) 2005a; Leucht 2005b). If data based on these thresholds are not between groups. We prefer not to calculate effect size measures available, we will use the primary cut-off presented by the original (standardised mean difference (SMD)). However, if scales of very authors. considerable similarity are used, we will presume there is a small difference in measurement and we will calculate effect size and 2.7 Direction of graphs transform the effect back to the units of one or more of the specific Where possible, we will enter data in such a way that the area instruments. to the left of the line of no effect indicates a favourable outcome for discontinuation of anticholinergics prescribed along with antipsychotics. Where keeping to this makes it impossible to avoid Unit of analysis issues outcome titles with clumsy double-negatives (e.g. ’Not un-improved’) we will report data where the left of the line indicates an unfavourable outcome and make a note in the relevant graphs. 1. Cluster trials Studies increasingly employ ’cluster randomisation’ (such as ran- Assessment of risk of bias in included studies domisation by clinician or practice) but analysis and pooling of Review authors SP and MA will work independently to assess risk clustered data poses problems. Firstly, authors often fail to account of bias by using criteria described in the Cochrane Handbook for for intra-class correlation in clustered studies, leading to a ’unit Systemic Reviews of Interventions to assess trial quality (Higgins of analysis’ error whereby P values are spuriously low, confidence 2011). This set of criteria is based on evidence of associations intervals unduly narrow and statistical significance overestimated between overestimate of effect and high risk of bias of the article (Divine 1992). This causes type I errors (Bland 1997; Gulliford such as sequence generation, allocation concealment, blinding, 1999). incomplete outcome data and selective reporting. Where clustering is not accounted for in primary studies, we will If the raters disagree, the final rating will be made by consensus. present data in a table, with a (*) symbol to indicate the presence Where inadequate details of randomisation and other characteris- of a probable unit of analysis error. In subsequent versions of this tics of trials are provided, we will contact authors of the studies in review we will seek to contact first authors of studies to obtain order to obtain further information. We will report non-concur- intra-class correlation coefficients for their clustered data and to rence in quality assessment; but if disputes arise as to which cate- adjust for this by using accepted methods (Gulliford 1999). Where gory a trial is to be allocated, we will again resolve by discussion. clustering has been incorporated into the analysis of primary stud- We will note the level of risk of bias in both the text of the review ies, we will present these data as if from a non-cluster randomised and in the ’Summary of findings’ table. study, but adjust 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 Measures of treatment effect effect’. This is calculated using the mean number of participants per cluster (m) and the intra-class correlation coefficient (ICC) (Design effect = 1 + (m 1) * ICC) (Donner 2002). If the ICC − 1. Binary data is not reported it will be assumed to be 0.1 (Ukoumunne 1999).

Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms (Protocol) 6 Copyright © 2017 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. If cluster studies have been appropriately analysed taking into ac- not. We will undertake a sensitivity analysis testing how prone the count intra-class correlation coefﬁcients and relevant data docu- primary outcomes are to change when data only from people who mented in the report, synthesis with other studies is possible using complete the study to that point are compared to the intention- the generic inverse variance technique. to-treat analysis using the above assumptions.

2. Cross-over trials 3. Continuous A major concern of cross-over trials is the carry-over effect. It oc- curs if an effect (e.g. pharmacological, physiological or psychological) of the treatment in the first phase is carried over to the 3.1 Attrition second phase. As a consequence, on entry to the second phase the In the case where attrition for a continuous outcome is between participants can differ systematically from their initial state despite 0% and 50%, and data only from people who complete the study a washout phase. For the same reason cross-over trials are not ap- to that point are reported, we will reproduce these. propriate if the condition of interest is unstable (Elbourne 2002). As both effects are very likely in severe mental illness, we will only 3.2 Standard deviations use data of the first phase of cross-over studies. If standard deviations are not reported, we will first try to obtain the missing values from the authors. If not available, where there 3. Studies with multiple treatment groups are missing measures of variance for continuous data, but an exact Where a study involves more than two treatment arms we will standard error and confidence intervals are available for group present the additional treatment arms, if relevant, in comparisons. means, and either P value or t value are available for differences in If data are binary we will simply add these and combine within the mean, we can calculate them according to the rules described in the two-by-two table. If data are continuous we will combine data fol- Cochrane Handbook for Systemic Reviews of Interventions (Higgins lowing the formula in section 7.7.3.8 (Combining groups) of the 2011). When only the standard error (SE) is reported, standard Cochrane Handbook for Systemic Reviews of Interventions (Higgins deviations (SDs) are calculated by the formula SD = SE * √(n). 2011). Where the additional treatment arms are not relevant, we Chapters 7.7.3 and 16.1.3 of the Cochrane Handbook for Systemic will not use these data. Reviews of Interventions present detailed formulae for estimating SDs from P values, t or F values, confidence intervals, ranges or other statistics (Higgins 2011). If these formulae do not apply, Dealing with missing data we will calculate the SDs according to a validated imputation method which is based on the SDs of the other included studies (Furukawa 2006). Although some of these imputation strategies 1. Overall loss of credibility can introduce error, the alternative would be to exclude a given At some degree of loss of follow-up, data must lose credibility (Xia study’s outcome and thus to lose information. We nevertheless will 2009). We choose that, for any particular outcome, should more examine the validity of the imputations in a sensitivity analysis than 50% of data be unaccounted for, we will not reproduce these excluding imputed values. data or use them within analyses. If, however, more than 50% of those in one arm of a study are lost, but the total loss is less than 50%, we will address this within the ’Summary of findings’ table/ 3.3 Assumptions about participants who left the trials early s by down-rating quality. Finally, we will also downgrade quality or were lost to follow-up within the ’Summary of findings’ table/s should loss be 25% to Various methods are available to account for participants who left 50% in total. the trials early or were lost to follow-up. Some trials just present the results of study completers, others use the method of last ob- servation carried forward (LOCF), while more recently methods 2. Binary such as multiple imputation or mixed effects models for repeated In the case where attrition for a binary outcome is between 0% and measurements (MMRM) have become more of a standard. While 50% and where these data are not clearly described, we will present the latter methods seem to be somewhat better than LOCF (Leon data on a ’once-randomised-always-analyse’ basis (an intention- 2006), we feel that the high percentage of participants leaving the to-treat analysis). Those leaving the study early are all assumed to studies early and differences in the reasons for leaving the stud- have the same rates of negative outcome as those who completed, ies early between groups is often the core problem in randomised with the exception of the outcome of death and adverse effects. schizophrenia trials. We will therefore not exclude studies based on For these outcomes we will use the rate of those who stay in the the statistical approach used. However, we will preferably use the study - in that particular arm of the trial - for those who did more sophisticated approaches: for example, we prefer MMRM

Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms (Protocol) 7 Copyright © 2017 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. or multiple-imputation to LOCF and will only present completer Assessment of reporting biases analyses if some kind of ITT data are not available at all. Moreover, we will address this issue in the item ’incomplete outcome data’ of the ’Risk of bias’ tool. 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 Assessment of heterogeneity described in section 10.1 of the Cochrane Handbook for Systemic Reviews of Interventions (Higgins 2011). We will try to locate protocols of included randomised trials. If the protocol is available, we will compare outcomes in the protocol and in the published 1. Clinical heterogeneity report. If the protocol is not available, we will compare outcomes We will consider all included studies initially, without seeing com- listed in the Methods section of the trial report with actually re- parison data, to judge clinical heterogeneity. We will simply in- ported results. spect all studies for clearly outlying people or situations which we had not predicted would arise. We will disucss such situations or participant groups when they arise. 2. Funnel plot Reporting biases arise when the dissemination of research findings is influenced by the nature and direction of results (Egger 1997). 2. Methodological heterogeneity These are again described in Section 10 of the Cochrane Handbook for Systemic Reviews of Interventions (Higgins 2011). We are aware We will consider all included studies initially, without seeing com- that funnel plots may be useful in investigating reporting biases parison data, to judge methodological heterogeneity. We will sim- but are of limited power to detect small-study effects. We will ply inspect all studies for clearly outlying methods which we had not use funnel plots for outcomes where there are 10 or fewer not predicted would arise. We will disucss such methodological studies, or where all studies are of similar size. In other cases, where outliers when they arise. funnel plots are possible, we will seek statistical advice in their interpretation.

3. Statistical heterogeneity Data synthesis We understand that there is no closed argument for preference for use of fixed-effect or random-effects models. The random-effects 3.1 Visual inspection method incorporates an assumption that the different studies are We will visually inspect graphs to investigate the possibility of estimating different, yet related, intervention effects. This often statistical heterogeneity. seems to be true to us and the random-effects model takes into account differences between studies even if there is no statistically significant heterogeneity. There is, however, a disadvantage to the random-effects model: it puts added weight onto small studies 3.2 Employing the I² statistic which often are the most biased ones. Depending on the direction We will investigate heterogeneity between studies by considering of effect these studies can either inflate or deflate the effect size. the I² method alongside the Chi² P value. The I² provides an es- We choose to use the random-effects model for all analyses. timate of the percentage of inconsistency thought to be due to chance (Higgins 2003). The importance of the observed value of I² depends on i. magnitude and direction of effects and ii. strength of Subgroup analysis and investigation of heterogeneity evidence for heterogeneity (e.g. P value from Chi² test, or a confidence interval for I²). We will interpret an I² estimate greater than or equal to 50% accompanied by a statistically significant Chi² 1. Subgroup analyses statistic as evidence of substantial levels of heterogeneity (Section

9.5.2 Higgins 2011). When substantial levels of heterogeneity are found in the primary outcome, we will explore reasons for het- 1.1 Primary outcomes erogeneity (Subgroup analysis and investigation of heterogeneity). We do not anticipate any sub-group analyses.

Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms (Protocol) 8 Copyright © 2017 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. 1.2 Clinical state, stage or problem Where assumptions have to be made regarding missing SDs data We propose to undertake this review and provide an overview of (see Dealing with missing data), we will compare the ﬁndings of the effects of discontinuation of anticholinergics prescribed along the primary outcomes when we use our assumption/s and when we with antipsychotics for people with schizophrenia in general. In use data only from people who complete the study to that point. addition, however, we aim to report, where available, data on sub- We will undertake a sensitivity analysis testing how prone results groups of people in the same clinical state, stage and with similar are to change when completer-only data only are compared to the problems. imputed data using the above assumption. If there is a substantial difference, we will report results and discuss them but will continue to employ our assumption. 2. Investigation of heterogeneity

We will report if inconsistency is high. First we will investigate 3. Risk of bias whether data have been entered correctly. Second, if data are cor- We will analyse the effects of excluding trials that are judged to be rect, we will visually inspect the graph and successively remove at high risk of bias across one or more of the domains of randomi- studies outside of the company of the rest until homogeneity is sation (implied as randomised with no further details available, restored. For this review we decided that should this occur with allocation concealment, blinding and outcome reporting) for the data contributing to the summary ﬁnding of no more than 10% meta-analysis of the primary outcome. If the exclusion of trials at of the total weighting, we will present these data. If not, we will high risk of bias does not substantially alter the direction of effect not pool data but discuss the issue instead. We know of no sup- or the precision of the effect estimates, then we will include data porting research for this 10% cut-off but are investigating use of from these trials in the analysis. prediction intervals as an alternative to this unsatisfactory state. When unanticipated clinical or methodological heterogeneity are obvious we will simply state hypotheses regarding these for future 4. Imputed values reviews or versions of this review.We do not anticipate undertaking We will also undertake a sensitivity analysis to assess the effects of analyses relating to these. including data from trials where we use imputed values for ICC in calculating the design effect in cluster randomised trials. Sensitivity analysis

If substantial differences are noted in the direction or precision of 5. Fixed and random effects effect estimates in any of the sensitivity analyses listed below, we We will synthesise data using a random-effects model; however, will not pool data from the excluded trials with the other trials we will also synthesise data for the primary outcome using a fixed- contributing data to the outcome, but will present them separately. effect model to evaluate whether this alters the significance of the result. 1. Implication of randomisation We aim to include trials in a sensitivity analysis if they are described in some way as to imply randomisation. For the primary outcomes ACKNOWLEDGEMENTS we will include these studies and if there is no substantive clinically significant difference when the implied randomised studies are The editorial base of Cochrane Schizophrenia produces and main- added to those with better description of randomisation, then we tains standard text for use in the Methods section of their reviews. will employ all data from these studies. We have used this text as the basis for what appears here and adapted it as required.

2. Assumptions for lost binary data The search term was developed by collaboration between the In- formation Specialist of Cochrane Schizophrenia and the contact Where assumptions have to be made regarding people lost to fol- author of this protocol. low-up (see Dealing with missing data) we will compare the ﬁnd- ings of the primary outcomes when we use our assumption/s and We would also like to thank and acknowledge Hongyong Deng when we use data only from people who complete the study to and Timothy Kariotis for peer reviewing this version of the proto- that point. If there is a substantial difference, we will report results col, and Rebecca Dickenson for her help with writing the protocol and discuss them but will continue to employ our assumption. by contributing from her previously published protocols

Additional references Donner 2002 Donner A, Klar N. Issues in the meta-analysis of cluster Altman 1996 randomized trials. Statistics in Medicine 2002;21(19): Altman DG, Bland JM. Detecting skewness from summary 2971–80. information. BMJ 1996;313(7066):1200. Egger 1997 American Psychiatric Association 2006 Egger M, Davey Smith G, Schneider M, Minder C. Bias American Psychiatric Association. American Psychiatric in meta-analysis detected by a simple, graphical test. BMJ Association Practice Guidelines for the Treatment of 1997;315(7109):629–34. PsychiatricDisorders: Compendium 2006. American Psychiatric Publishing, 2006. Elbourne 2002 Bland 1997 Elbourne D, Altman DG, Higgins JPT, Curtina F, Bland JM. Statistics notes. Trials randomised in clusters. Worthingtond HV, Vaile A. Meta-analyses involving cross- BMJ 1997;315(7108):600. over trials: methodological issues. International Journal of Epidemiology 2002;31(1):140–9. BNF 2015 British National Formulary. www.medicinescomplete.com/ Furukawa 2006 mc/bnf/current/PHP3134-antimuscarinic-drugs-used- Furukawa TA, Barbui C, Cipriani A, Brambilla P,Watanabe in-parkinsonism.htm. BMJ Group and the Royal N. Imputing missing standard deviations in meta-analyses Pharmaceutical Society of Great Britain, August 2015. can provide accurate results. Journal of Clinical Epidemiology Boissel 1999 2006;59(7):7–10. Boissel JP, Cucherat M, Li W, Chatellier G, Gueyffier F, Gervin 2000 Buyse M, et al. The problem of therapeutic efficacy indices. Gervin M, Barnes TRE. Assessment of drug-related 3. Comparison of the indices and their use [Apercu sur movement disorders in schizophrenia. Advances in la problematique des indices d’efficacite therapeutique, 3: Psychiatric Treatment 2000;6(5):332–41. comparaison des indices et utilisation. Groupe d’Etude des Indices D’efficacite]. Therapie 1999;54(4):405–11. Gulliford 1999 [PUBMED: 10667106] Gulliford MC. Components of variance and intraclass correlations for the design of community-based surveys Bottiggi 2006 and intervention studies: data from the Health Survey for Bottiggi KA, Salazar JC, Yu L, Caban-Holt AM, Ryan England 1994. American Journal of Epidemiology 1999;149 M, Mendiondo MS, et al. Long-term cognitive impact (9):876–83. of anticholinergic medications in older adults. American Journal of Geriatric Psychiatry 2006;14(11):980–4. Higgins 2003 Buhrich 2000 Higgins JP, Thompson SG, Deeks JJ, Altman DG. Buhrich N, Weller A, Kevans P. Misuse of anticholinergic Measuring inconsistency in meta-analyses. BMJ 2003;327 drugs by people with serious mental illness. Psychiatric (7414):557–60. Services 2000;51(7):928–9. Higgins 2011 Deeks 2000 Higgins JPT, Green S (editors). Cochrane Handbook Deeks J. Issues in the selection for meta-analyses of binary for Systematic Reviews of Interventions Version 5.0.2 data. 8th International Cochrane Colloquium; 2000 [updated March 2011]. The Cochrane Collaboration, Oct 25-28; Cape Town. Cape Town: The Cochrane 2011. Available from handbook.cochrane.org. Collaboration, 2000. Hutton 2009 Dickenson 2014a Hutton JL. Number needed to treat and number needed to Dickenson R, Momcilovic S, Donnelly L. Anticholinergics harm are not the best way to report and assess the results versus placebo for neuroleptic-induced parkinsonism. of randomised clinical trials. British Journal of Haematology Cochrane Database of Systematic Reviews 2014, Issue 6. 2009;146(1):27–30. [DOI: 10.1002/14651858.CD011164] Dickenson 2014b Johnstone 1983 Dickenson R, Bethi SK, Srivastava S. Anticholinergics Johnstone EC, Crow TJ, Ferrier IN, Frith CD, Owens (various) for neuroleptic-induced parkinsonism. Cochrane DGC, Bourne RC, et al. Adverse effects of anticholinergic Database of Systematic Reviews 2014, Issue 5. [DOI: medication on positive schizophrenic symptoms. 10.1002/14651858.CD011113] Psychological Medicine 1983;13(03):513–27. Divine 1992 Kay 1986 Divine GW, Brown JT, Frazier LM. The unit of analysis Kay SR, Opler LA, Fiszbein A. Positive and Negative error in studies about physicians’ patient care behavior. Syndrome Scale (PANSS) Manual. North Tonawanda, NY: Journal of General Internal Medicine 1992;7(6):623–9. Multi-Health Systems, 1986.

Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms (Protocol) 10 Copyright © 2017 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Leon 2006 Schünemann 2008 Leon AC, Mallinckrodt CH, Chuang-Stein C, Archibald Schünemann HJ, Oxman AD, Vist GE, Higgins JPT, Deeks DG, Archer GE, Chartier K. Attrition in randomized JJ, Glasziou P, et al. Chapter 12: Interpreting results and controlled clinical trials: methodological issues in drawing conclusions. In: Higgins JPT, Green S editor(s). psychopharmacology. Biological Psychiatry 2006;59(11): Cochrane Handbook for Systematic Reviews of Interventions. 1001–5. [PUBMED: 16905632] Chichester (UK): John Wiley & Sons, 2008:359–83.

Leucht 2005a Soares-Weiser 1997 Leucht S, Kane JM, Kissling W, Hamann J, Etschel E, Soares-Weiser K, Mobsy C, Holliday E. Anticholinergic Engel R. Clinical implications of brief psychiatric rating medication for neuroleptic-induced tardive dyskinesia. scale scores. British Journal of Psychiatry 2005;187:366–71. Cochrane Database of Systematic Reviews 1997, Issue 2. [PUBMED: 16199797] [DOI: 10.1002/14651858.CD000204] Leucht 2005b Leucht S, Kane JM, Kissling W, Hamann J, Etschel E, Stahl 2008 Engel RR. What does the PANSS mean?. Schizophrenia Stahl SM. Stahl’s Essential Psychopharmacology: Research 2005;79(2-3):231–8. [PUBMED: 15982856] Neuroscientiﬁc Basis and Practical Applications. Cambridge University Press, 2008. Marshall 2000 Marshall M, Lockwood A, Bradley C, Adams C, Joy C, Taylor 2012 Fenton M. Unpublished rating scales: a major source Taylor D, Paton C, Kapur S. The Maudsley Prescribing of bias in randomised controlled trials of treatments for Guidelines. Informa healthcare, 2012. schizophrenia. British Journal of Psychiatry 2000;176: Ukoumunne 1999 249–52. Ukoumunne OC, Gulliford MC, Chinn S, Sterne JAC, Novick 2010 Burney PGJ. Methods for evaluating area-wide and Novick D, Haro JM, Bertsch J, Haddad PM. Incidence organistation-based intervention in health and health care: of extrapyramidal symptoms and tardive dyskinesia in a systematic review. Health Technology Assessment 1999;3(5): schizophrenia: thirty-six-month results from the European 1–75. Schizophrenia outpatient health outcomes study. Journal of Clinical Psychopharmacology October 2010;30(5):531–40. WHO 2015 World Health Organisation. WHO Model List of Overall 1962 Essential Medicines. www.who.int/medicines/publications/ Overall JE, Gorham DR. The brief psychiatric rating scale. essentialmedicines/EML2015_8-May-15.pdf May 2015, Psychological Reports 1962;10:799–812. issue 19:25,38–39. Pullen 1984 Pullen GP,Best NR, Maguire J. Anticholinergic drug abuse: Xia 2009 a common problem?. British Medical Journal 1984;289 Xia J, Adams CE, Bhagat N, Bhagat V, Bhoopathi P, El- (6445):612–3. Sayeh H, et al. Loss to outcomes stakeholder survey: the LOSS study. Psychiatric Bulletin 2009;33(7):254–7. Rathbone 2006 Rathbone J, Soares-Weiser K. Anticholinergics for Yassa 1988 neuroleptic-induced acute akathisia. Cochrane Database Yassa, R. Tardive dyskinesia and anticholinergic drugs. A of Systematic Reviews 2006, Issue 4. [DOI: 10.1002/ critical review of the literature. Encéphale 1988;14:233–9. 14651858.CD003727.pub3] ∗ Indicates the major publication for the study

CONTRIBUTIONSOFAUTHORS

Simi Pereira writing the protocol Michael Albert - help with writing protocol

Anticholinergic discontinuation for antipsychotic-induced extra-pyramidal symptoms (Protocol) 11 Copyright © 2017 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. DECLARATIONSOFINTEREST

Simi Pereira none known Michael Albert - none known.

SOURCES OF SUPPORT

Internal sources Nottinghamshire Healthcare NHS Trust, UK. • Employer of Rebecca Dickenson Norfolk and Suffolk NHS Foundation Trust, UK. • Employer of Simi Pereira and Michael Albert

External sources No sources of support supplied •