Number Needed To Treat Statistic Research Articles

Quantifying the Protective Effects of Stimulants on Functional Outcomes in Attention-Deficit/Hyperactivity Disorder: A Focus on Number Needed to Treat Statistic and Sex Effects

PurposeThe aim of the study was to help quantify the protective effects of stimulant treatment on important functional outcomes in attention-deficit/hyperactivity disorder (ADHD) using the number needed to treat (NNT) statistic and examine whether these effects are moderated by sex. MethodsSubjects were derived from three independent samples, two similarly designed case–control, 10-year prospective follow-up studies of boys and girls with and without ADHD grown up and a cross-sectional randomized clinical trial of lisdexamfetamine on driving performance and behavior. For all studies, subjects were evaluated with structured diagnostic interviews. To measure psychopharmacologic treatment in the follow-up studies, we collected information about each subject's stimulant medication use, age at onset, and age at termination of treatment. Subjects in the driving study underwent two driving simulation assessments (premedication and after 6 weeks of treatment on lisdexamfetamine or placebo). For each outcome, we ran a logistic regression model that included an interaction between sex and treatment status. Lifetime rates were used to calculate the NNT statistic. We also calculated adjusted NNT statistics that accounted for sex, age, socioeconomic status, and family intactness. ResultsThe NNTs were very low, ranging from 3 to 10. No interaction effects with sex were detected (all p > .05). The adjusted NNTs mostly remained the same with the exception of any substance use disorder, which increased after controlling for age. ConclusionsStimulants have strong protective effects on functional outcomes in youth with ADHD that are not moderated by sex. These results support the critical importance of early identification and treatment of children with ADHD of both sexes.

Bromperidol decanoate (depot) for schizophrenia

Antipsychotic drugs are the mainstay treatment for schizophrenia. Long-acting depot injections of drugs such as bromperidol decanoate are extensively used as a means of long-term maintenance treatment. To assess the effects of depot bromperidol versus placebo, oral antipsychotics and other depot antipsychotic preparations for people with schizophrenia in terms of clinical, social and economic outcomes. For this 2012 update we searched the Cochrane Schizophrenia Group's Register (February 2012). We sought all randomised trials focusing on people with schizophrenia where depot bromperidol, oral antipsychotics or other depot preparations. Primary outcomes were clinically significant change in global function, service utilisation outcomes (hospital admission, days in hospital), relapse. For the 2011 update MP independently extracted data, CEA carried out the reliability check. We calculated fixed-effect risk ratios (RR) and 95% confidence intervals (CI) for dichotomous data, and calculated weighted or standardised means for continuous data. Where possible, we calculated the number needed to treat statistic (NNT). Analysis was by intention-to-treat.For the 2012 update, data collection and analysis was not carried out as no new studies were found. The 2012 search found no new studies, we have therefore included no new trials in this 2012 update. The number of included trials remain 4 RCTs, total n = 117. A single, small study of six months' duration compared bromperidol decanoate with placebo injection. Similar numbers left the study before completion (n = 20, 1 RCT, RR 0.4 CI 0.1 to 1.6) and there were no clear differences between bromperidol decanoate and placebo for a list of adverse effects (n = 20, 1 RCT, RR akathisia 2.0 CI 0.21 to 18.69, RR increased weight 3.0 CI 0.14 to 65.9, RR tremor 0.33 CI 0.04 to 2.69). When bromperidol decanoate was compared with fluphenazine depot, we found no important change on global outcome (n = 30, RR no clinical important improvement 1.50 CI 0.29 to 7.73). People allocated to fluphenazine decanoate and haloperidol decanoate had fewer relapses than those given bromperidol decanoate (n = 77, RR 3.92 Cl 1.05 to 14.60, NNH 6 CI 2 to 341). People allocated bromperidol decanoate required additional antipsychotic medication somewhat more frequently than those taking fluphenazine decanoate and haloperidol decanoate, but the results did not reach conventional levels of statistical significance (n = 77, 2 RCTs, RR 1.72 CI 0.7 to 4.2). The use of benzodiazepine drugs was very similar in both groups (n = 77, 2 RCTs, RR 1.08 CI 0.68 to 1.70). People left the bromperidol decanoate group more frequent than those taking other depot preparation due to any cause (n = 97, 3 RCTs, RR 2.17 CI 1.00 to 4.73). Anticholinergic adverse effects were equally common between bromperidol and other depots (n = 47, RR 3.13 CI 0.7 to 14.0) and additional anticholinergic medication was needed with equal frequency in both depot groups, although results did tend to favour the bromperidol decanoate group (n = 97, 3 RCTs, RR 0.80 CI 0.64 to 1.01). The incidence of movement disorders was similar in both depot groups (n = 77, 2 RCTs, RR 0.74 CI 0.47 to 1.17). Minimal poorly reported trial data suggests that bromperidol decanoate may be better than placebo injection but less valuable than fluphenazine or haloperidol decanoate. If bromperidol decanoate is available it may be a viable choice, especially when there are reasons not to use fluphenazine or haloperidol decanoate. Well-conducted and reported randomised trials are needed to inform practice.

Depot fluspirilene for schizophrenia.

Antipsychotic drugs are the mainstay treatment for schizophrenia and similar psychotic disorders. Long-acting depot injections of drugs such as fluspirilene are extensively used as a means of long-term maintenance treatment. To review the effects of depot fluspirilene versus placebo, oral anti-psychotics and other depot antipsychotic preparations for people with schizophrenia in terms of clinical, social and economic outcomes. We searched the Cochrane Schizophrenia Group's Register (September 2005), inspected references of all identified studies, and contacted relevant pharmaceutical companies. We included all relevant randomised trials focusing on people with schizophrenia where depot fluspirilene, oral anti-psychotics, other depot preparations, or placebo were compared. Outcomes such as death, clinically significant change in global function, mental state, relapse, hospital admission, adverse effects and acceptability of treatment were sought. Studies were reliably selected, quality rated and data extracted. For dichotomous data, we calculated relative risk (RR) with the 95% confidence intervals (CI). Where possible, the number needed to treat statistic (NNT) was calculated. Analysis was by intention-to-treat. We summated normal continuous data using the weighted mean difference (WMD). We presented scale data only for those tools that had attained pre-specified levels of quality. We included twelve randomised studies in this update of which five are additional studies. One trial compared fluspirilene and placebo and did not report important differences in the global improvement (n=60, 1 RCT, RR "no important improvement "0.97 CI 0.9 to 1.1). Though movement disorders (n=60, 1 RCT, RR 31.0 CI 1.9 to 495.6, NNH 4) were found only in the fluspirilene group, there were no convincing data showing the advantage of oral chlorpromazine or other depot antipsychotics over fluspirilene decanoate. We found no difference between depot fluspirilene and other oral antipsychotics with regard to relapses or to the number of people leaving the study early. Global state data (CGI) were not significantly different, in the short term when comparing fluspirilene with other depots (n=90, 2 RCTs, RR "no important improvement" 0.80 CI 0.2 to 2.8). No significant difference were apparent between fluspirilene and other depots with respect to the number of people leaving the trial early (n=83, 2 RCTs, RR 0.55 CI 0.1 to 2.3) or relapse rates (n=109, 3 RCTs, RR 0.55 CI 0.1 to 2.3). Extrapyramidal adverse effects were significantly less prevalent in the fluspirilene groups (n=164, 4 RCTs, RR 0.50 CI 0.3 to 0.8, NNH 5). Other adverse effects were not significantly different. Attrition in the one comparison between fluspirilene in weekly versus biweekly administration (n=34, RR 3.00 CI 0.1 to 68.8) and relapse rates (n=34 RR 3.18 CI 0.1 to 83.8) were not significantly different. There were no significant difference for movement disorders in one short term study. No study reported on hospital and service outcomes or commented on participants' overall satisfaction with care. Economic outcomes were not recorded by any of the included studies. Participant numbers in each comparison were small and we found no clear differences between fluspirilene and oral medication or other depots. The choice of whether to use fluspirilene as a depot medication and whether it has advantages over other depots cannot, at present, be informed by trial-derived data. Well-conducted and reported randomised trials are still needed to inform practice.

Day centres for severe mental illness

The number of people with severe mental illness who receive treatment whilst living at home has increased greatly over the last 30 years. Day centres and day hospitals frequently supplement this treatment. To determine the effects of non-medical day centre care for people with severe mental illness. We updated our search in September 2005. All databases and searches are detailed in the body of the text. We would have included all randomised controlled trials where seriously mentally ill people were allocated to non-medical day centre care. We reliably selected studies, quality rated them and extracted data. For dichotomous data, it had been hoped to estimate the fixed effects Relative Risk (OR) with 95% confidence intervals (CI) and the number needed to treat statistic (NNT). Analysis was to have been by intention-to-treat. Normal continuous data were to have been summated using the weighted mean difference (WMD) and scale data presented only for those tools that had attained pre-specified levels of quality. Electronic searches identified over 300 citations but none were relevant to this review. We found no trials of non-medical day centres. We feel that the inclusion of any studies less rigorous than randomised trials would result in misleading findings and that it is not unreasonable to expect well designed, conducted and reported randomised controlled trials of day centre care. More precise nomenclature would greatly help identify relevant work. At present non-randomised comparative studies give conflicting messages about the roles provided by day centres and the clinical and social needs they are able to meet. It is therefore probably best that people with serious mental illness and their carers, if given the choice, take a pragmatic decision on which type of unit best meets their needs. There is a clear need for randomised controlled trials of day centre care compared to other forms of day care, and when resources are limited, day centre care within the context of a pragmatic randomised trial may be the only way of ensuring equity of provision.

Supported housing for people with severe mental disorders

There has been a significant reduction in the number of people with severe mental illness who spend extended periods in long-stay hospitals. District health authorities, local authorities, housing associations and voluntary organisations are jointly expected to provide support for people with severe mental disorder/s. This 'support' may well involve some kind of special housing. To determine the effects of supported housing schemes compared with outreach support schemes or 'standard care' for people with severe mental disorder/s living in the community. For the 2006 update we searched the Cochrane Schizophrenia Group Trials Register (April 2006) and the Cochrane Central Register of Controlled Trials (CENTRAL, 2006 Issue 2). We included all relevant randomised, or quasi-randomised, trials dealing with people with 'severe mental disorder/s' allocated to supported housing, compared with outreach support schemes or standard care. We focused on outcomes of service utilisation, mental state, satisfaction with care, social functioning, quality of life and economic data. We reliably selected studies, quality rated them and undertook data extraction. For dichotomous data, we would have estimated relative risks (RR), with the 95% confidence intervals (CI). Where possible, we would have calculated the number needed to treat statistic (NNT). We would have carried out analysis by intention-to-treat and would have summated normal continuous data using the weighted mean difference (WMD). We would have presented scale data for only those tools that had attained pre-specified levels of quality and undertaken tests for heterogeneity and publication bias. Although 139 citations were acquired from the searches, no study met the inclusion criteria. Dedicated schemes whereby people with severe mental illness are located within one site or building with assistance from professional workers have potential for great benefit as they provide a 'safe haven' for people in need of stability and support. This, however, may be at the risk of increasing dependence on professionals and prolonging exclusion from the community. Whether or not the benefits outweigh the risks can only be a matter of opinion in the absence of reliable evidence. There is an urgent need to investigate the effects of supported housing on people with severe mental illness within a randomised trial.

An under-used yet easily understood statistic: the number needed to treat (NNT)

Objectives Interpretation of the research literature can be difficult, yet it is essential in order for physiotherapists to convey the details of risks and benefits associated with therapeutic interventions to patients. The number needed to treat (NNT) statistic is one such statistic that is easily interpretable. Using several examples from the physiotherapy literature representing a range of conditions, this article demonstrates how to convert more commonly used statistics into the NNT statistic. The purpose of this is to show how the NNT can help clinicians to converse with patients to convey details about the likelihood of benefit with treatment and/or the likelihood of risk, in order that a decision may be made with respect to therapy. Design The Australian Physiotherapy Evidence Database (PEDro) was searched in order to locate a selection of physiotherapy research articles that reported various dichotomous outcomes that could be converted to the NNT statistic for the purpose of this analysis. Results The NNT statistic for nine studies with a PEDro score ≥6 was calculated using the Internet-based downloadable spreadsheet on the PEDro website. For six studies, the NNT point estimates ranged from 2 to 4 (95% confidence interval 1–10). One study had a NNT of 8, while two other studies produced number needed to harm values. Conclusion The NNT can be calculated quickly and efficiently using Internet-based calculators and/or other decision-making tools, and may be an alternative that provides readily interpretable information to assist in conveying the likely benefits (and/or risks) of treatment to patients.

Annualized was found better than absolute risk reduction in the calculation of number needed to treat in chronic conditions

Background and Objective Recent studies have calculated number needed to treat (NNT) estimates based on annualized rates; however, the ramifications of altering the NNT statistic have not yet been explored in the literature. Here we introduce the concept of annualized NNT (ANNT), and apply it to data from randomized controlled trials (RCTs). Methods Incidence rates from RCTs for serious adverse events for three medicines were compared to an older class of drugs. NNT and ANNT were calculated from the event rates for these events. Results Based on the data, the NNT to prevent one adverse event a year vs. older medications was drug A, ANNT = 88; drug B, ANNT = 77; drug C, ANNT = 68. Equivalent calculations based on Bayesian statistics are drug C, ANNT = 54; drug B, ANNT = 49. Drug A produced a bimodal distribution, with one mode within the NNT range and the other in the number needed to harm range. Conclusions NNT can erroneously inflate differences between treatments when based on absolute and not differential safety. We propose that NNT be limited to acute conditions with short-term, well-defined treatment courses, and that ANNT be used for chronic conditions.

Depot perphenazine decanoate and enanthate for schizophrenia

Antipsychotic drugs are usually given orally but compliance with medication given by this route may be difficult to quantify. The development of depot injections in the 1960s gave rise to extensive use of depots as a means of long-term maintenance treatment. Perphenazine decanoate and enanthate are depot antipsychotics that belong to the phenothiazine family and have a piperazine ethanol side chain. To assess the effects of depot perphenazine decanoate and enanthate versus placebo, oral antipsychotics and other depot antipsychotic preparations for people with schizophrenia in terms of clinical, social and economic outcomes. We updated previous searches of the Cochrane Schizophrenia Group Register (June 1998), Biological Abstracts (1982-1998), the Cochrane Library (Issue 2, 1998), EMBASE (1980-1998), MEDLINE (1966-1998), and PsycLIT (1974-1998) by searching the Cochrane Schizophrenia Group Register (March 2004). References of all identified trials were also inspected for more studies and industry contacted. We compared randomised clinical trials focusing on people with schizophrenia where depot perphenazine decanoate and enanthate, oral antipsychotics or other depot preparations. We reliably selected studies, quality rated them and extracted data. For dichotomous data we estimated the Relative Risk (RR) with the 95% confidence intervals (CI). Where possible, we calculated the number needed to treat statistic (NNT). Analysis was by intention-to-treat. Only four studies (Ahlfors 1980, Eufe 1979, Knudsen 1985c, Tegeler 1979), randomising a total 313 people could be included in this review and this combined with an overall lack of usable data limits any interpretation of results. Perphenazine enanthate was not significantly any better or worse than other depot antipsychotics in most of the main outcomes such as global state, relapse or leaving the study early. We found some differences favouring the control groups for adverse effects. One study (Ahlfors 1980) of six months' duration (n=172), compared perphenazine enanthate to clopenthixol decanoate. There were no differences between the two groups for outcomes of global improvement, relapse and leaving the study early. More people in the perphenazine enanthate group, however, required anticholinergic drugs than those allocated to clopenthixol decanoate (RR 1.12 CI 1.0 to 1.2, NNT 10).A single study (n=64, duration six weeks) compared perphenazine enanthate and its longer acting decanoate ester. Data on relapse and leaving the study early failed to show convincing differences. The enanthate group, however, experienced more movement disorders (RR 1.36, CI 1.1 to 1.8 NNT 5) than those allocated the decanoate ester of the same drug and required more anticholinergic drugs (RR 1.47 CI 1.1 to 2.0, NNT 4). Depot perphenazine is in clinical use in the Nordic countries, Belgium, Portugal and the Netherlands. At a conservative estimate, a quarter of a million people suffer from schizophrenia in those countries and could be treated with depot perphenazine. The total number of participants in the four trials with useful data is 313. None of the studies observed the effects of oral versus depot antipsychotic drugs. Until well conducted and reported randomised trials are undertaken clinicians will be in doubt as to the effects of perphenazine depots and people with schizophrenia should exercise their own judgement or ask to be randomised.

Using the number needed to treat in clinical practice

Weeks DL, Noteboom JT. Using the number needed to treat in clinical practice. Arch Phys Med Rehabil 2004;85:1729–31. The number needed to treat (NNT) is gaining attention as a method of reporting the results of clinical trails with dichotomous outcome measures. The NNT is defined as the number of patients who would need to be treated, on average, with a specific intervention to prevent 1 additional bad outcome or to achieve 1 desirable outcome in a given time period. Because it reports outcomes in terms of patient numbers, it is extremely useful to clinicians for making decisions about the effort expended with a particular intervention to achieve a single positive outcome. This special communication describes the NNT statistic and its utility for choosing clinical interventions.

Depot bromperidol decanoate for schizophrenia.

Antipsychotic drugs are the mainstay treatment for schizophrenia. Long-acting depot injections of drugs such as bromperidol decanoate are extensively used as a means of long-term maintenance treatment. To assess the effects of depot bromperidol versus placebo, oral antipsychotics and other depot antipsychotic preparations for people with schizophrenia in terms of clinical, social and economic outcomes. Relevant trials were identified by searching Biological Abstracts (1982-1999), Cochrane Library (Issue 2, 1999), Cochrane Schizophrenia Group's Register (May 1999), EMBASE (1980-1999), MEDLINE (1966-1999) and PsycLIT (1974-1999). References of all identified trials were inspected and Janssen-Cilag was contacted in order to identify more trials. An update search was undertaken in October 2003. The Schizophrenia Groups trials register is based on regular searches of BIOSIS Inside; CENTRAL; CINAHL; EMBASE; MEDLINE and PsycINFO; the hand searching of relevant journals and conference proceedings, and searches of several key grey literature sources. A full description is given in the Group's module. All randomised trials focusing on people with schizophrenia where depot bromperidol, oral antipsychotics or other depot preparations were sought. Primary outcomes were death, clinically significant change in global function, mental state, relapse, hospital admission, adverse effects and acceptability of treatment. Data were extracted independently by two reviewers and cross-checked. Fixed effects relative risks (RR) and 95% confidence intervals (CI) were calculated for dichotomous data. Weighted or standardised means were calculated for continuous data. Where possible, the number needed to treat statistic (NNT) was calculated. Analysis was by intention-to-treat. Four controlled clinical trials were included (total n=117). We identified a single small study of six months duration comparing bromperidol decanoate with placebo injection. Similar numbers left the study before completion (n=20, 1 RCT, RR 0.4 CI 0.1 to 1.6) and there was no clear differences between bromperidol decanoate and placebo for a list of adverse effects (n=20, 1 RCT, RR akathisia 2.0 CI 0.21 to 18.69, RR increased weight 3.0 CI 0.14 to 65.9, RR tremor 0.33 CI 0.04 to 2.69). When bromperidol decanoate was compared with fluphenazine depot we found no important change on global outcome (n=30, RR no clinical important improvement 1.50 CI 0.29 to 7.73). People allocated to fluphenazine decanoate and haloperidol decanoate had less relapses than those given bromperidol decanoate (n=77, RR 3.92 Cl 1.05 to 14.60, NNH 6 CI 2 to 341). People allocated bromperidol decanoate required additional antipsychotic medication somewhat more frequently than those taking fluphenazine decanoate and haloperidol decanoate but the results did not reach conventional levels of statistical significance (n=77, 2 RCTs, RR 1.72 CI 0.7 to 4.2). The use of benzodiazepine drugs was very similar in both groups (n=77, 2 RCTs, RR 1.08 CI 0.68 to 1.70). People left the bromperidol decanoate group with the same frequency as those allocated other depots (n=97, 3 RCTs, RR 1.92 CI 0.8 to 4.6). Anticholinergic adverse effects were equally common between bromperidol and other depots (n=47, RR 3.13 CI 0.7 to 14.0) and additional anticholinergic medication was needed with equal frequency in both depot groups, although results did tend to favour the bromperidol decanoate group (n=97, 3 RCTs, RR 0.80 CI 0.64 to 1.01). The incidence of movement disorders was similar in both depot groups (n=77, 2 RCTs, RR 0.74 CI 0.47 to 1.17). Currently, minimal poorly reported trial data suggests that bromperidol decanoate may be better than placebo injection but less valuable than fluphenazine or haloperidol decanoate. If bromperidol decanoate is available it may be a viable choice, especially when there are reasons not to use fluphenazine or haloperidol decanoate. Well-conducted and reported randomised trials are needed to inform practice in Belgium, Germany, Italy and the Netherlands.

Supported housing for people with severe mental disorders.

There has been a significant reduction in the number of people with severe mental illness who spend extended periods in long-stay hospitals. Psychiatric and social services, both statutory and voluntary, aim to assist these people to stay in their local community. District health authorities, local authorities, housing associations and voluntary organisations are jointly expected to provide support for people with severe mental disorder/s. This 'support' may well involve some sort of special housing. To determine the effects of supported housing schemes compared with outreach support schemes or 'standard care' for people with severe mental disorder/s living in the community. Cochrane Schizophrenia Group's Register of trials (February 2001) and the Cochrane Library (Issue 1, 2001) were searched using relevant phrases. These databases are compiled by methodical searches of BIOSIS, CINAHL, Dissertation abstracts, EMBASE, LILACS, MEDLINE, PSYNDEX, PsycINFO, RUSSMED, Sociofile, supplemented with hand searching of relevant journals and numerous conference proceedings. Reference list screening of relevant papers was performed. Relevant randomised, or quasi-randomised, trials dealing with people with 'severe mental disorder/s' allocated to supported housing, outreach support schemes or standard care focusing on outcomes of service utilisation, mental state, satisfaction with care, social functioning, quality of life, and economic data, were sought. Studies were reliably selected, quality rated and data extracted. For dichotomous data, relative risks (RR) would have been estimated, with the 95% confidence intervals (CI). Where possible, the number needed to treat statistic (NNT) was to have been calculated. Analysis would have been by intention-to-treat. Normal continuous data were to have been summated using the weighted mean difference (WMD). Scale data were to have been presented for only those tools that had attained pre-specified levels of quality. Tests of heterogeneity and for publication bias were to have been undertaken. No studies met the inclusion criteria although 139 citations were acquired from the searches. Dedicated schemes whereby people with severe mental illness are located within one site or building with assistance from professional workers have potential for great benefit as they provide a 'safe haven' for people in need of stability and support. This, however, may be at the risk of increasing dependence on professionals and prolonging exclusion from the community. Whether or not the benefits outweigh the risks can only be a matter of opinion in the absence of reliable evidence. There is an urgent need to investigate the effects of supported housing on people with severe mental illness within a randomised trial.

Cognitive behaviour therapy for schizophrenia.

Medication is the mainstay of treatment for schizophrenia. Many people with schizophrenia, however, continue to experience symptoms in spite of medication and may experience side effects that are unwanted and unpleasant. In addition to medication additional forms of treatment include talking therapies such as cognitive behavioural therapy. This approach helps to link the person's feelings and patterns of thinking which underpin distress. To review the effectiveness of cognitive behavioural therapy for people with schizophrenia, when compared to standard care, specific medication, other therapies and non-intervention. Electronic searches of Biological Abstracts (1980-1998), CINAHL (1982-1998), The Cochrane Library (Issue 2, 1998), The Cochrane Schizophrenia Groups' Register of Trials, which encompasses up to date searches of all listed databases (January 2001), EMBASE (1980-1998), MEDLINE (1966-1998), PsychLIT (1887-1997), SIGLE (1990-1998), Sociofile (1980-1998) were undertaken. All references of the articles selected were searched for further relevant trials. This review includes relevant randomised trials of cognitive behaviour therapy for people with a diagnosis of schizophrenia-like illnesses. Outcomes such as death, mental state, relapse, psychological well-being and acceptability of treatment were sought. Studies were reliably selected and assessed for methodological quality. Data were extracted by two reviewers working independently. Dichotomous data were analysed on an intention-to-treat basis and continuous data with 65% completion rate are presented. Where possible, for dichotomous outcomes, a relative risk (RR) with the 95% confidence interval (CI) was estimated along with the number needed to treat statistic (NNT). Twenty-two relevant papers describing thirteen trials were identified. Cognitive behavioural therapy in addition to standard care did not significantly reduce the rate of relapse and readmission to hospital when compared with standard care alone (medium term 1 RCT, N=61, RR 0.1 CI 0.01 to 1.7; long term 2 RCTs, N=123, RR 1.1 CI 0.8 to 1.5). A significant difference was observed, however, favouring cognitive behavioural therapy over standard care alone, in terms of being able to be discharged from hospital (1 RCT, N=62, RR 0.5 CI 0.3 to 0.9, NNT 3 CI 2 to 12). For 'no important improvement in mental state' data showed a significant difference favouring the cognitive behavioural therapy group over standard care alone when measured at 13 to 26 weeks (2 RCTs, N=123, RR 0.7 CI 0.6 to 0.9, NNT 4 CI 2 to 8). After one year the difference was no longer significant (3 RCTs, N=211, RR 0.95 CI 0.6 to 1.5). On continuous measures (BPRS, CPRS, Psychiatric Assessment Scale) data are not convincing of an effect. A cognitive behavioural therapy approach focusing on compliance may have some effects on insight and attitudes to medication, but the clinical meaning of these data is unclear. When compared with supportive psychotherapy, cognitive behavioural therapy had no effects on relapse rate and clinically meaningful improvements in mental state. Cognitive behavioural therapy combined with other psycho-social/educational interventions may decrease the numbers of people able to tolerate the intervention, at least under study conditions. Cognitive behavioural therapy is a promising but under evaluated intervention. Currently, trial-based data supporting the wide use of cognitive behavioural therapy for people with schizophrenia or other psychotic illnesses are far from conclusive. More trials are justified, especially in comparison with a lower grade supportive approach. These trials should be designed to be both clinically meaningful and widely applicable.

'As required' medication regimens for seriously mentally ill people in hospital.

Drugs used to treat psychotic illnesses may take weeks to have an effect. Often, in the interim, additional, 'as required' doses of medication can be used to calm patients in psychiatric wards. 20 to 50% of people on acute psychiatric wards are written up for 'as required' doses of medication. In this situation, a doctor prescribes the frequency and upper limit of dose, and the drug is then given at the discretion of clinical staff. To compare the effects of 'as required' medication regimens with regular regimens of medication for the treatment of psychotic symptoms or behavioural disturbance, thought to be secondary to psychotic illness. This is in addition to any regular psychotropic medication for the long-term treatment of schizophrenia or schizophrenia-like illnesses. The reviewers searched the Cochrane Schizophrenia Group's register of trials (November 2001). This register is compiled by methodical searches of BIOSIS, CINAHL, Dissertation abstracts, EMBASE, LILACS, MEDLINE, PSYNDEX, PsycINFO, RUSSMED, Sociofile and is supplemented with hand searching of relevant journals and numerous conference proceedings. Relevant randomised control trials, involving hospital inpatients with schizophrenia or schizophrenia-like illnesses, comparing any regimen of medication administered for the short term relief of behavioural disturbance, or psychotic symptoms, to be given at the discretion of ward staff ('as required', 'prn') with fixed non-discretionary patterns of drug administration of the same drug(s). This is in addition to regular psychotropic medication for the long-term treatment of schizophrenia or schizophrenia-like illnesses where prescribed. The primary outcomes of interest were levels of disturbance, severity of psychiatric symptoms, dose of medication, and adverse effects. Studies were reliably selected, quality rated and data extracted. For dichotomous data, relative risks (RR) would have been estimated, with the 95% confidence intervals (CI). Where possible, the number needed to treat statistic (NNT) was to have been calculated. Analysis would have been by intention-to-treat. Normal continuous data were to have been summated using the weighted mean difference (WMD). Scale data were to have been presented for only those tools which had attained pre-specified levels of quality. Tests of heterogeneity and for publication bias were to have been undertaken. No randomised trials comparing 'as required' medication regimens to regular regimens of the same drug were identified. This common current practice has no support from randomised trials. Current practice is based on clinical experience and habit rather than high quality evidence. Current practice, therefore, outside of a well designed, conducted and reported randomised trial, is therefore difficult to justify.

Depot pipothiazine palmitate and undeclynate for schizophrenia.

Anti-psychotic drugs are usually given orally but compliance with medication given by this route may be difficult to quantify. The development of depot injections in the 1960s gave rise to extensive use of depots as a means of long-term maintenance treatment. Pipothiazine palmitate is a depot from the phenothiazine family of antipsychotic drugs. To assess the effects of depot pipothiazine palmitate and undeclynate versus placebo, oral anti-psychotics and other depot antipsychotic preparations for people with schizophrenia in terms of clinical, social and economic outcomes. Relevant trials were identified by searching Biological Abstracts (1982-1998), Cochrane Library (Issue 2, 1998), Cochrane Schizophrenia Group's Register (June 1998), EMBASE (1980-1998), MEDLINE (1966-1998) and PsycLIT (1974-1998). References of all identified trials were also inspected for more studies and industry contacted. All clinical randomised trials focusing on people with schizophrenia where depot pipothiazine palmitate and undeclynate, oral anti-psychotics or other depot preparations were compared. Studies were reliably selected, quality rated and data extracted. For dichotomous data Peto odds ratios (OR) with the 95% confidence intervals (CI) were estimated. Where possible, the number needed to treat statistic (NNT) was calculated. Analysis was by intention-to-treat. Normal continuous data were summated using the weighted mean difference (WMD). Scale data were presented only for those tools that had attained pre-specified levels of quality. Fourteen studies were included. When pipothiazine palmitate was compared to 'standard' oral antipsychotics no differences were found for outcomes of mental state, study attrition, behaviour and adverse effects (total randomised = 166). Pipothiazine palmitate was compared to other depot preparations in nine studies (n=455). Again no differences were identified for outcomes of global improvement, mental state, study attrition, behaviour and adverse effects. Although well-conducted and reported randomised trials are still needed to inform practice (no trial data exists reporting hospital and services outcomes, satisfaction with care and economics) pipothiazine palmitate is a viable choice for clinician and recipient of care. Data suggests it is not different to other depot antipsychotics.

Day centres for severe mental illness.

The number of people with severe mental illness who receive treatment whilst living at home has increased greatly over the last 30 years. Day centres and day hospitals frequently supplement this treatment. To determine the effects of non-medical day centre care for people with severe mental illness. The Allied and Complementary Medicine Database (1985-1999), The British Nursing Index (1994-1998), The Cochrane Library (Issue 2, 1999), The Cochrane Schizophrenia Group's Register (May 1999), EMBASE (1980-1999), MEDLINE (1966-1999), PsycLIT (1887-1999), The Royal College of Nurses Database (1985-1996), and Sociological Abstracts (1963-1999) were searched. References of all identified studies were also inspected for more studies. All randomised controlled trials where seriously mentally ill people were allocated to non-medical day centre care. Studies were reliably selected, quality rated and data extracted. For dichotomous data, it had been hoped to estimate the Peto odds ratios (OR) with 95% confidence intervals (CI) and the number needed to treat statistic (NNT). Analysis was to have been by intention-to-treat. Normal continuous data were to have been summated using the weighted mean difference (WMD) and scale data presented only for those tools that had attained pre-specified levels of quality. Despite extensive searching, no trials were found of non-medical day centres. The electronic search identified over 300 citations but none was relevant to this review. The reviewers feel that the inclusion of any studies less rigorous than randomised trials would result in misleading findings and that it is not unreasonable to expect well designed, conducted and reported randomised controlled trials of day centre care. More precise nomenclature would greatly help identify relevant work. At present non-randomised comparative studies give conflicting messages about the roles provided by day centres and the clinical and social needs they are able to meet. It is therefore probably best that people with serious mental illness and their carers, if given the choice, take a pragmatic decision on which type of unit best meets their needs. There is a clear need for randomised controlled trials of day centre care compared to other forms of day care, such as day hospitals.

Clotiapine for acute psychotic illnesses.

Acute psychotic illness, especially when associated with agitated or violent behaviour, requires urgent pharmacological tranquillisation or sedation. Clotiapine, a dibenzothiazepine neuroleptic, is being used for this purpose in several countries. To estimate the effects of clotiapine when compared to other 'standard' or 'non-standard' treatments of acute psychotic illness in controlling disturbed behaviour and reducing psychotic symptoms. The Cochrane Controlled Trials Register (Issue 2, 2000), The Cochrane Schizophrenia Group's Register (May 2000), EMBASE (1980-2000), MEDLINE (1966-2000), PASCAL (1973-2000) and PsycLIT (1970-2000) were methodically searched. This was supplemented by hand searching reference lists, contacting industry and relevant authors. Randomised clinical trials comparing clotiapine to any treatment, for people with acute psychotic illnesses such as in schizophrenia, schizoaffective disorder, mixed affective disorders, manic phase of bipolar disorder, brief psychotic episode or organic psychosis following substance abuse. Studies were reliably selected, quality assessed and data extracted. Data were excluded where more than 50% of participants in any group were lost to follow up. For binary outcomes a standard estimation of the risk ratio (RR) and its 95% confidence interval (CI) was calculated. Where possible, the weighted number needed to treat statistic (NNT), and its 95% confidence interval (CI), was also calculated. If heterogeneity was found, a random effects model was used. For continuous outcomes, endpoint data were preferred to change data. Non-skewed data from valid scales were summated using a weighted mean difference (WMD). Again, if heterogeneity was found a random effects model was used. A Mantel-Haenszel chi-square test was used to investigate the possibility of heterogeneity. Five trials were included. None compared clotiapine with placebo, but control drugs were either antipsychotics (chlorpromazine, perphenazine, trifluoperazine and zuclopenthixol acetate) or benzodiazepines (lorazepam). Versus antipsychotics: results for global clinical outcome were heterogeneous (p=0.09) but did not suggest clotiapine to be superior, or inferior, to chlorpromazine, perphenazine, or trifluoperazine (total randomised = 83). Use of clotiapine did change the proportion of people ready for hospital discharge by the end of the study in one small trial (n=49, RR 1.04 95%CI 0.96 to 2.12). Overall, attrition rates were low. No significant difference was found for those allocated to clotiapine compared with people randomised to other antipsychotics (n=121, RR 2.26 95%CI 0.40 to 13). Weak data suggests that clotiapine may result in less need for antiparkinsonian treatment compared with zuclopenthixol acetate (n=38, RR 0.43 95%CI 0.02 to 0.98). Versus lorazepam: when used to control aggressive/violent outbursts for people already treated with haloperidol, clotiapine did not significantly improve mental state compared to lorazepam (WMD -3.36 95%CI -8.09 to 1.37). Much data could not be pooled due to skew or inadequate presentation of results. Economic outcomes and satisfaction with care were not addressed. We found no significant evidence to support the use of clotiapine rather than other 'standard' or 'non-standard' treatments for the management of acute psychotic illness. The trials included in this review all present important methodological flaws. We do not wish to discourage clinicians from using clotiapine in the psychiatric emergency, we would just like to point out the fact that good quality controlled trials are needed on this subject.

Pimozide for schizophrenia or related psychoses.

Pimozide was first formulated in the late 1960s and continues to be marketed for the care of those with schizophrenia or related psychoses such as delusional disorder. Pimozide is generally well tolerated apart from extrapyramidal side effects. It has, however, been associated with cardiotoxicity and sudden unexplained deaths and electrocardiogram monitoring is now required before and during its use. To assess the effects of pimozide for people with schizophrenia, non-affective psychotic mental illness and delusional disorder in terms of clinical, social and economic outcomes. Electronic searches of Biological Abstracts (1982-1995), The Cochrane Schizophrenia Group's Register, EMBASE (1980-1995), Janssen-Cilag UK's register of studies (1999), MEDLINE (1966-1995), PsycLIT (1974-1995), hand-searching the references of all included studies and contacting the manufacturers of the compound. All randomised trials relating to people with schizophrenia, or similar disorders comparing pimozide to other drug treatments were sought. Studies where randomisation was implied rather than stated were included if they did not change the results. Primary outcomes were clinically significant change in global function, mental state, relapse, hospital admission, death, adverse events and acceptability of treatment. Studies were selected, rated and data extracted. For dichotomous data Relative Risks (RR) based on a random effects model with 95% confidence intervals (CI) were estimated. The number needed to treat statistic (NNT) was calculated where indicated. Analysis was by intention-to-treat. This review currently includes 34 studies focusing on those with schizophrenia, none on people with delusional disorder. Few people have been randomised to pimozide versus placebo. Data from two longer term studies does suggest that the active drug prevents relapse (n=66, RR 0.45 CI 0.24-0.86, NNT 4, CI 3-22) but the confidence interval is wide. Pimozide has similar efficacy to that of typical antipsychotic drugs for the outcomes of change in global functioning, mental state, relapse and leaving the study early. People allocated to pimozide did not have a higher mortality than those taking other antipsychotics. Pimozide was more likely to cause parkinsonian tremor (RR 1.6 CI 1.1-2.3, NNH 6 CI 3-44) and lead to a requirement for antiparkinsonian medication more frequently (RR 1.8, CI 1.2-2.6, NNH 3 CI 2-5) than other drugs. It was, however, less likely to cause sedation (RR 0.38 CI 0.2-0.7, NNH 6 CI 4-16). Although there are shortcomings in the data there is enough overall consistency, over different outcomes and time scales, to confirm that pimozide is a drug with similar efficacy to other more commonly used antipsychotics such as chlorpromazine for those with schizophrenia. There are no data to support or refute its use for those with delusional disorder.

Token economy for schizophrenia.

A token economy is a behavioural therapy technique in which the desired change is achieved by means of tokens administered for the performance of predefined behaviours according to a program. Though token economy programmes were widespread in the 1970s they became largely restricted to wards where long-stay patients from institutions are prepared for transfer into the community and were particularly aimed at changing negative symptoms of schizophrenia - poor motivation, poor attention and social withdrawal. To review the effects of token economies for people with schizophrenia, or other serious or chronic mental illnesses, compared with standard care. Electronic searches of Biological Abstracts (1985-1999), CINAHL (1982-1998), The Cochrane Library (Issue 1, 1999), The Cochrane Schizophrenia Group's Register of Trials (February 1999), EMBASE (1980-1999) and PsycLIT (1987-1998) were supplemented with reference searches, personal contact with trial authors and hand searches. Randomised studies comparing a token economy regime (one in which change is achieved by means of use of non-monetary, non-consumable tokens) to standard care for those with schizophrenia or any other similar chronic or serious mental illness. Studies were reliably selected, quality rated and data extracted. For dichotomous data relative risk (RR) with 95% confidence intervals (CI) was estimated. Where possible, the number needed to treat statistic (NNT) was also calculated. Analysis was by intention-to-treat. Normal continuous data were summated using the weighted mean difference (WMD). Scale data were presented only for those tools that had attained pre-specified levels of quality. Only three randomised controlled trials could be included in the analyses (total n=110). There were no usable data on target or non-target behaviour. One small study favoured the token economy approach for the outcome 'change in mental state' on the SANS-CV with improvement in negative symptoms at three months (n=40, WMD -12. 7, CI -21.44 to -3.96). The token economy approach may have effects on negative symptoms but it is unclear if these results are reproducible, clinically meaningful and are maintained beyond the treatment programme. Token economy remains worthy of careful evaluation in well designed, conducted and reported randomised trials.

Depot perphenazine decanoate and enanthate for schizophrenia.

Anti-psychotic drugs are usually given orally but compliance with medication given by this route may be difficult to quantify. The development of depot injections in the 1960s gave rise to extensive use of depots as a means of long-term maintenance treatment. Perphenazine decanoate and enanthate are depot antipsychotics that belong to the phenothiazine family and have a piperazine ethanol side chain. To assess the effects of depot perphenazine decanoate and enanthate versus placebo, oral anti-psychotics and other depot antipsychotic preparations for people with schizophrenia in terms of clinical, social and economic outcomes. Biological Abstracts (1982-1998), the Cochrane Library (Issue 2, 1998), the Cochrane Schizophrenia Group's Register (June 1998), EMBASE (1980-1998), MEDLINE (1966-1998), and PsycLIT (1974-1998) were searched. References of all identified trials were also inspected for more studies and industry contacted. Randomised clinical trials focusing on people with schizophrenia where depot perphenazine decanoate and enanthate, oral anti-psychotics or other depot preparations were compared. Studies were reliably selected, quality rated and data extracted. For dichotomous data Peto odds ratios (OR) with the 95% confidence intervals (CI) were estimated. Where possible, the number needed to treat statistic (NNT) was calculated. Analysis was by intention-to-treat. One study of six months duration, compared perphenazine enanthate to clopenthixol decanoate. There was no differences between the two for outcomes of global improvement, relapse and leaving the study early. More people in the perphenazine enanthate group required anticholinergic drugs than those allocated to clopenthixol decanoate (OR 3.6 CI 1.2-10, NNT 10). A single study (n=64, duration six weeks) compared perphenazine enanthate and its longer acting decanoate ester. Data on relapse and leaving the study early failed to show convincing differences. The enanthate group, however, experienced more movement disorders (OR 0.2 CI 0.06-0.7) than those allocated the decanoate ester of the same drug (NNT 4.0) and required more anticholinergic drugs (OR 0.2 CI 0.08-0.7, NNT 3.7). Depot perphenazine is in clinical use in the Nordic countries, Belgium, Portugal and the Netherlands. At a conservative estimate a quarter of a million people suffer from schizophrenia in those countries and could be treated with depot perphenazine. The total number of participants in the two trials with useful data is 236. Neither study observes the effect of oral versus depot antipsychotic drugs. Until well conducted and reported randomised trials are undertaken clinicians will be in doubt as to the effects of perphenazine depots and people with schizophrenia should exercise their own judgement or ask to be randomised.

Depot bromperidol decanoate for schizophrenia.

Anti-psychotic drugs are the mainstay treatment for schizophrenia. Long-acting depot injections of drugs such as bromperidol decanoate are extensively used as a means of long-term maintenance treatment. To assess the effects of depot bromperidol versus placebo, oral anti-psychotics and other depot antipsychotic preparations for people with schizophrenia in terms of clinical, social and economic outcomes. Relevant trials were identified by searching Biological Abstracts (1982-1999), Cochrane Library (Issue 2, 1999), Cochrane Schizophrenia Group's Register (May 1999), EMBASE (1980-1999), MEDLINE (1966-1999) and PsycLIT (1974-1999). References of all identified trials were inspected and Janssen Cilag contacted in order to identify more trials. All randomised trials focusing on people with schizophrenia where depots bromperidol, oral anti-psychotics or other depot preparations were sought. Primary outcomes were death, clinically significant change in global function, mental state, relapse, hospital admission, adverse effects and acceptability of treatment. Studies were reliably selected, quality rated and data extracted. For dichotomous data Peto odds ratios (OR) with the 95% confidence intervals (CI) were estimated. The number needed to treat statistic (NNT) was to have been calculated. Analysis was by intention-to-treat. Four controlled clinical trials were found (total n=117). Smeraldi 1990 (n=20) compared bromperidol decanoate to placebo and found that more people in the latter group left the study by six months duration (50% versus 20%, OR 0.3 CI 0.05-7). There were no clear differences between bromperidol decanoate and placebo for a list of side effects. Ratings of global impression, mental state and needing additional antipsychotic medication all tended to favour the control depots (fluphenazine decanoate and haloperidol decanoate) and people consistently left the bromperidol decanoate group more frequently than those allocated other depots (n=97, OR 2.6 CI 0.8-9). There was no clear pattern in the occurrence of adverse effects. Currently, extrapolating from minimal trial data suggests that bromperidol decanoate may be better than a placebo injection but less valuable than fluphenazine or haloperidol decanoate. If bromperidol decanoate is available to the clinician it may be a viable choice, especially when there are reasons not to use fluphenazine or haloperidol decanoate. Well-conducted and reported randomised trials are urgently needed to inform practice in Belgium, Germany, Italy and the Netherlands.