Date created: | Last Updated:

Category: Analysis

Description: Review question To systematically assess the clinical efficacy and safety of long-acting injectable second-generation antipsychotics compared to their oral formulations or placebo for people with acute schizophrenia. Searches The following sources will be searched without restrictions for language or publication period: 1. Cochrane Schizophrenia Group’s Study-Based Register(we will search Cochrane Schizophrenia Group’s Study-Based Register). This register is compiled by regular searches in the following databases. 2. Excerpta Medica Database (EMBASE) 3. Cumulative Index to Nursing and Allied Health Literature (CINAHL) 4. American Psychological Association (PsycINFO) 5. PubMed 6. US National Institute of Health Ongoing Trials Register (ClinicalTrials.gov) 7. World Health Organization International Clinical Trials Registry Platform (www. who. int/ictrp) Searching other resources 1. Cited reference searching We will inspect the references of all identified trials for other published reports and citations of unpublished studies 2. Search of other systematic reviews We will also check previously published relevant systematic reviews to check if some studies meet our inclusion criteria. 3. Personal contact We will contact the primary authors of all studies initially selected for inclusion and the responsible drug companies if there is some missing information. Types of study to be included We will include randomized controlled trials (RCTs) in which patients with schizophrenia received an intervention as defined below (see Types of interventions section). We will accept open, single, and double-blinded trials. In the case of cross-over studies, the first cross-over phase will be included to avoid the problem of carryover effects which are very likely in drugs for schizophrenia (Elbourne, et al. 2002). We will accept studies with the duration of treatment over three weeks; this choice is mainly related to there is evidence from previous trials that a substantial effect of antipsychotics needs at least three weeks. No language restriction will be in the present study to minimize “language bias”(Egger, et al. 1997). We will exclude studies from mainland China for which major quality concerns have been raised(Woodhead 2016), except for studies conducted by international pharmaceutical companies in mainland China . Condition or domain being studied Schizophrenia and schizophrenia-like disorders (schizoaffective disorder, schizophreniform disorder, delusion disorder). Participants/population We will include individuals diagnosed with schizophrenia and related disorders (at least 80%) in an acute phase by any criteria irrespective of gender, age, or race. There is no clear, unified diagnostic standard for the acute phase of schizophrenia. To select this population, we operationalized the inclusion criteria as follows. We considered people to be in the acute phase of schizophrenia if they were experiencing an exacerbation in their baseline level of symptoms or if they had active symptoms and were currently hospitalized. In addition, if the authors described the patients in the study as acute or did not mention that patients were stable, we will assume them as acute. We will exclude maintenance studies in stable patients (relapse prevention studies) or dose reduction studies. Intervention(s), exposure(s) We only include SGA LAI given in 2 weeks or longer intervals. To the best of our knowledge, the SGAs with licensed LAI formulations include aripiprazole, olanzapine, risperidone, and paliperidone, because other oral antipsychotics do not exist by injection. We will exclude FGA LAI because these studies are older. After all, the clinical applications of FGA LAI are less common than SGA LAI (Janzen, et al. 2020). In addition, a recent article has revealed that older studies showed larger intervention effects than that of recent RCTs; thus, the authors suggest that meta-analyses including older trials should be interpreted cautiously(Smail-Faugeron, et al. 2022). Type of comparators 1. Olanzapine LAI (any dose) versus olanzapine oral or placebo 2. Paliperidone LAI (any dose) versus paliperidone oral or placebo 3. Risperidone LAI (any dose) versus risperidone oral or placebo 4. Aripiprazole LAI (any dose) versus aripiprazole oral or placebo 5. Any second-generation LAI (any dose) from the four antipsychotics in 1.-4.vs oral antipsychotics (always same counterpart to the LAIs) or placebo. 6. The four antipsychotics in 1.-4. as oral compound vs. placebo. These comparisons shall provide an estimate of the effect with oral compounds, which can then be compared to the effect with LAI applications in 1.-4. The studies comparing a SG LAI with its oral formulation (same compound) or placebo will be included in the review. We will exclude studies comparing a SG LAI to a different oral drug (different compound) or a different LAI. Context There are no restrictions in terms of setting, for example, we will include in- and outpatients. Types of outcome measures Primary outcome 1. Change in overall symptoms, measured by rating scales such as the PANSS (Kay, et al. 1987) or the BPRS (Overall and Gorham 1988) total score, or any other published scale (e.g., the Manchester Scale) to assess overall schizophrenic symptomatology. The results of other rating scales will only be used if the instrument has been published in a peer-reviewed journal. Secondary outcomes 1. Response to treatment. We will accept the original author’s decisions, but if different options are available we will prefer rating scale defined criteria in the following hierarchy: At least 50% reduction on PANSS or BPRS, CGI (Guy 1976) much improved, <50% to >20% reduction on PANSS or BPRS, 20% reduction on PANSS or BPRS, CGI minimally improved. 2. Change in PANSS positive scale score. 3. Change in PANSS negative scale score 4. Dropout due to any reason. 5. Dropout due to specific reasons. Dropout due to inefficacy of treatment will be considered as an additional outcome of the efficacy of treatment. Dropout due to the occurrence of adverse events will be used as a measure of overall tolerability. 6. Depression, measured by the Calgary Depression Scale for Schizophrenia, the Hamilton Depression Rating Scale(Williams 1988), the Montgomery Asberg Depression Scale(Davidson, et al. 1986), or other published symptom scales. 7. Quality of life, measured by any published rating scale (e.g Quality of Life Scale (Heinrichs, et al. 1984) 8. Functioning, measured by any published rating scale (e.g. global assessment of functioning (Hall 1995), Personal and Social Performance scale PSP REF). 9. Mortality: we will examine this outcome in terms of (A) death for any reason, (B) death due to natural causes, and (C) due to suicide. 10. The following major side effects will be examined: using antiparkinson drugs as a measure of extrapyramidal side-effects akathisia, weight gain in kg, number of participants with 7% weight gain or more, prolactin levels, sedation or somnolence, QTc prolongation, and at least one anticholinergic side-effect. Measures of effect 1. Continuous outcomes: We will use the standardized mean difference (SMD) between two groups and its 95% confidence intervals (CIs) if some studies employ different scales. Nevertheless, we will use mean differences (MD) for weight gain (kg), prolactin levels (ng/ml) and QTc prolongation (ms), since we can convert values of these outcomes into the same metric. The trials may report the results either as endpoint means or using changes in mean values from the baseline assessment. We will give preference to the mean change from baseline to endpoint measures, and, if not available, we will take the mean values at the endpoint. 2. Dichotomous outcomes: The effect size for dichotomous outcomes will be odds ratios (OR) and 95% confidence intervals (CIs), because the odds ratio has better mathematical properties than the relative risk. But we will convert back to relative risks (RRs) and percentages in treatment and control groups for presentation of the results. Dose‐ranging studies If a study has multiple arms with the same medication administered at different doses or administered for a different time length, we will pool these intervention groups into a single one, as recommended by the Cochrane Handbook for Systematic Reviews of Interventions, section 16.5.4 (Higgins, et al. 2019). In fixed dose studies, only those doses which are listed in the summary of product characteristics of the drugs will be included. We will include all doses from flexible dose studies in which physicians can adapt the dose to the individual patient. Dealing with missing data To some extent, missing outcome data is common, influencing credibility (Xia, et al. 2009). Among the methods that attempt to take attrition into account, we will prefer more sophisticated approaches such as multiple imputations or mixed-effects models (MMRM) to simple last-observation carried forward. Completer analyses will be included only if no other data are available. Moreover, we will address this issue in the' Incomplete outcome data' item of the RoB 2. If some studies did not report SDs, we would calculate SDs from reporting statistics, e.g., SE (Higgins, et al. 2019). If these are not available, we will contact authors, and if there is no reply, we will impute them from the other studies (Higgins, et al. 2019). Data extraction (selection and coding) Two authors will independently inspect the titles and abstracts from the search results. Full texts of included references will be obtained and independently accessed by two authors for eligible studies. Discrepancies will be resolved by discussion and SL will be involved if needed. Again, two authors will independently extract data into Access in duplicate. They will compare the two copies and resolve discrepancies. When they cannot reach consensus, SL will help clarify the issues, and these final decisions will be recorded. Where necessary, we will attempt to contact the author via open-ended request to obtain missing information or clarification. Risk of bias (quality) Two review authors will evaluate the risk of bias using Cochrane risk-of-bias tool for randomized trials (RoB 2) (Sterne, et al. 2019) for included LAI studies (comparisons 1.-4.). Discrepancies will be resolved by discussion and SL will be involved if needed. Strategy for data synthesis 1. we will do pairwise meta-analysis with random‐effects model inverse variance models . The random‐effects method incorporates an assumption that the different studies are estimating different, yet related, intervention effects. We feel that in schizophrenia which is a very heterogeneous disorder this is likely to be more appropriate. For dichotomous outcomes with rare Events we will use fixed effects Mantel-Haenszel models as recommended by Efthimiou et (Efthimiou 2018). because heterogeneity cannot be estimated well in the presence of rare events. 2. Assessment of heterogeneity: We will initially consider all included studies to judge methodological heterogeneity without seeing comparison data. We will simply inspect all studies for clearly outlying methods that we had not predicted would arise and discuss any such methodological outliers. In addition, we plan to investigate heterogeneity by visual inspection of forest plots and estimating the I2 statistic. Analysis of subgroups or subsets The following potential effect moderators of the primary outcome will be explored by subgroup analyses: 1. Different kinds of LAI formulations with same compounds, for example, there are two kinds of Aripiprazole LAI, Aripiprazole monohydrate (AM) and Aripiprazole lauroxil (AL). 2. Different dosing schedules for the same LAI. For example, olanzapine could be monthly or two weekly. 3. First-episode vs non-first-episode study populations. Sensitivity analyses of the primary outcome will be performed as follow: 1. Exclusion of non-double-blind studies (open and single-blind studies) 2. Exclusion of studies that presented only completer analyses 3. Exclusion of studies with an overall assessment of high risk of bias 4. Time point of outcome measurement in studies of LAIs vs placebo close to the Primary time point used in acute phase studies of oral compounds vs placebo (i.e. comparison 6), which is typically 6-8 weeks, whereas the LAI-studies typically have a primary time point of outcome measurement of => 12 weeks. References: Davidson, Jonathan, et al. 1986 The Montgomery‐Åsberg Depression Scale: reliability and validity. Acta psychiatrica scandinavica 73(5):544-548. Efthimiou, Orestis 2018 Practical guide to the meta-analysis of rare events. Evidence-based mental health 21(2):72-76. Egger, Matthias, et al. 1997 Language bias in randomised controlled trials published in English and German. The Lancet 350(9074):326-329. Elbourne, Diana R, et al. 2002 Meta-analyses involving cross-over trials: methodological issues. International journal of epidemiology 31(1):140-149. Guy, WBRR 1976 CGI. Clinical global impressions. ECDEU assessment manual for psychopharmacology. Hall, Richard CW 1995 Global assessment of functioning: a modified scale. Psychosomatics 36(3):267-275. Heinrichs, Douglas W, Thomas E Hanlon, and William T Carpenter Jr 1984 The Quality of Life Scale: an instrument for rating the schizophrenic deficit syndrome. Schizophrenia bulletin 10(3):388-398. Higgins, Julian PT, et al. 2019 Cochrane handbook for systematic reviews of interventions: John Wiley & Sons. Janzen, Donica, et al. 2020 Trends in the use of long-acting injectable antipsychotics in the province of Manitoba, Canada. Journal of Clinical Psychopharmacology 40(1):6-13. Kay, Stanley R, Abraham Fiszbein, and Lewis A Opler 1987 The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophrenia bulletin 13(2):261-276. Overall, John E, and Donald R Gorham 1988 The Brief Psychiatric Rating Scale (BPRS): recent developments in ascertainment and scaling. Psychopharmacology bulletin. Smail-Faugeron, Violaine, et al. 2022 Meta-analyses frequently include old trials that are associated with a larger intervention effect: a meta-epidemiological study. Journal of Clinical Epidemiology. Sterne, Jonathan AC, et al. 2019 RoB 2: a revised tool for assessing risk of bias in randomised trials. bmj 366. Williams, Janet BW 1988 A structured interview guide for the Hamilton Depression Rating Scale. Archives of general psychiatry 45(8):742-747. Woodhead, Michael 2016 80% of China’s clinical trial data are fraudulent, investigation finds: British Medical Journal Publishing Group. Xia, Jun, et al. 2009 Losing participants before the trial ends erodes credibility of findings. Psychiatric Bulletin 33(7):254-257.