To adjust or not to adjust in living systematic reviews? It's all about the context.

With each update of meta-analyses from living systematic reviews, treatment effects and their confidence intervals are recalculated. This often raises the question whether or not multiplicity is an issue and whether a method to adjust for multiplicity is needed. It seems that answering these questions is not that straightforward. We approach this matter by considering the context of systematic reviews and pointing out existing methods for handling multiplicity in meta-analysis. We conclude that multiplicity is not a relevant issue in living systematic reviews when they are planned with the aim to provide up-to-date evidence, without any direct control on the decision over future research. Multiplicity might be an issue, though, in living systematic reviews designed under a protocol involving a "stopping decision", which can be the case in living guideline development or in reimbursement decisions. Several appropriate methods exist for handling multiplicity in meta-analysis. Existing methods, however, are also associated with several technical and conceptual limitations, and could be improved in future methodological projects. To better decide whether an adjustment for multiplicity is necessary at all, authors and users of living systematic reviews should be aware of the context of the work and question whether there is a dependency between the effect estimates of the living systematic review and its stopping/updating or an influence on future research.

LFK index does not reliably detect small-study effects in meta-analysis: A simulation study.

The LFK index has been promoted as an improved method to detect bias in meta-analysis. Putatively, its performance does not depend on the number of studies in the meta-analysis. We conducted a simulation study, comparing the LFK index test to three standard tests for funnel plot asymmetry in settings with smaller or larger group sample sizes. In general, false positive rates of the LFK index test markedly depended on the number and size of studies as well as the between-study heterogeneity with values between 0% and almost 30%. Egger's test adhered well to the pre-specified significance level of 5% under homogeneity, but was too liberal (smaller groups) or conservative (larger groups) under heterogeneity. The rank test was too conservative for most simulation scenarios. The Thompson-Sharp test was too conservative under homogeneity, but adhered well to the significance level in case of heterogeneity. The true positive rate of the LFK index test was only larger compared with classic tests if the false positive rate was inflated. The power of classic tests was similar or larger than the LFK index test if the false positive rate of the LFK index test was used as significance level for the classic tests. Under ideal conditions, the false positive rate of the LFK index test markedly and unpredictably depends on the number and sample size of studies as well as the extent of between-study heterogeneity. The LFK index test in its current implementation should not be used to assess funnel plot asymmetry in meta-analysis.

When does the placebo effect have an impact on network meta-analysis results?

The placebo effect is the 'effect of the simulation of treatment that occurs due to a participant's belief or expectation that a treatment is effective'. Although the effect might be of little importance for some conditions, it can have a great role in others, mostly when the evaluated symptoms are subjective. Several characteristics that include informed consent, number of arms in a study, the occurrence of adverse events and quality of blinding may influence response to placebo and possibly bias the results of randomised controlled trials. Such a bias is inherited in systematic reviews of evidence and their quantitative components, pairwise meta-analysis (when two treatments are compared) and network meta-analysis (when more than two treatments are compared). In this paper, we aim to provide red flags as to when a placebo effect is likely to bias pairwise and network meta-analysis treatment effects. The classic paradigm has been that placebo-controlled randomised trials are focused on estimating the treatment effect. However, the magnitude of placebo effect itself may also in some instances be of interest and has also lately received attention. We use component network meta-analysis to estimate placebo effects. We apply these methods to a published network meta-analysis, examining the relative effectiveness of four psychotherapies and four control treatments for depression in 123 studies.

Model selection for component network meta-analysis in connected and disconnected networks: a simulation study.

BACKGROUND: Network meta-analysis (NMA) allows estimating and ranking the effects of several interventions for a clinical condition. Component network meta-analysis (CNMA) is an extension of NMA which considers the individual components of multicomponent interventions. CNMA allows to "reconnect" a disconnected network with common components in subnetworks. An additive CNMA assumes that component effects are additive. This assumption can be relaxed by including interaction terms in the CNMA. METHODS: We evaluate a forward model selection strategy for component network meta-analysis to relax the additivity assumption that can be used in connected or disconnected networks. In addition, we describe a procedure to create disconnected networks in order to evaluate the properties of the model selection in connected and disconnected networks. We apply the methods to simulated data and a Cochrane review on interventions for postoperative nausea and vomiting in adults after general anaesthesia. Model performance is compared using average mean squared errors and coverage probabilities. RESULTS: CNMA models provide good performance for connected networks and can be an alternative to standard NMA if additivity holds. For disconnected networks, we recommend to use additive CNMA only if strong clinical arguments for additivity exist. CONCLUSIONS: CNMA methods are feasible for connected networks but questionable for disconnected networks.

Pharmacological interventions for pruritus in adult palliative care patients.

BACKGROUND: This is the second update of the original Cochrane review published in 2013 (issue 6), which was updated in 2016 (issue 11). Pruritus occurs in patients with disparate underlying diseases and is caused by different pathologic mechanisms. In palliative care patients, pruritus is not the most prevalent but is a burdening symptom. It can cause considerable discomfort and negatively affect patients' quality of life. OBJECTIVES: To assess the effects of different pharmacological treatments compared with active control or placebo for preventing or treating pruritus in adult palliative care patients. SEARCH METHODS: For this update, we searched CENTRAL (the Cochrane Library), MEDLINE (OVID) and Embase (OVID) up to 6 July 2022. In addition, we searched trial registries and checked the reference lists of all relevant studies, key textbooks, reviews and websites, and we contacted investigators and specialists in pruritus and palliative care regarding unpublished data. SELECTION CRITERIA: We included randomised controlled trials (RCTs) assessing the effects of different pharmacological treatments, compared with a placebo, no treatment, or an alternative treatment, for preventing or treating pruritus in palliative care patients. DATA COLLECTION AND ANALYSIS: Two review authors independently assessed the identified titles and abstracts, performed data extraction and assessed the risk of bias and methodological quality. We summarised the results descriptively and quantitatively (meta-analyses) according to the different pharmacological interventions and the diseases associated with pruritus. We assessed the evidence using GRADE and created 13 summary of findings tables. MAIN RESULTS: In total, we included 91 studies and 4652 participants in the review. We added 42 studies with 2839 participants for this update. Altogether, we included 51 different treatments for pruritus in four different patient groups. The overall risk of bias profile was heterogeneous and ranged from high to low risk. The main reason for giving a high risk of bias rating was a small sample size (fewer than 50 participants per treatment arm). Seventy-nine of 91 studies (87%) had fewer than 50 participants per treatment arm. Eight (9%) studies had low risk of bias in the specified key domains; the remaining studies had an unclear risk of bias (70 studies, 77%) or a high risk of bias (13 studies, 14%). Using GRADE criteria, we judged that the certainty of evidence for the primary outcome (i.e. pruritus) was high for kappa-opioid agonists compared to placebo and moderate for GABA-analogues compared to placebo. Certainty of evidence was low for naltrexone, fish-oil/omega-3 fatty acids, topical capsaicin, ondansetron and zinc sulphate compared to placebo and gabapentin compared to pregabalin, and very low for cromolyn sodium, paroxetine, montelukast, flumecinol, and rifampicin compared to placebo. We downgraded the certainty of the evidence mainly due to serious study limitations regarding risk of bias, imprecision, and inconsistency. For participants suffering from uraemic pruritus (UP; also known as chronic kidney disease (CKD)-associated pruritus (CKD-aP)), treatment with GABA-analogues compared to placebo likely resulted in a large reduction of pruritus (visual analogue scale (VAS) 0 to 10 cm): mean difference (MD) -5.10, 95% confidence interval (CI) -5.56 to -4.55; five RCTs, N = 297, certainty of evidence: moderate. Treatment with kappa-opioid receptor agonists (difelikefalin, nalbuphine, nalfurafine) compared to placebo reduced pruritus slightly (VAS 0 to 10 cm, MD -0.96, 95% CI -1.22 to -0.71; six RCTs, N = 1292, certainty of evidence: high); thus, this treatment was less effective than GABA-analogues. Treatment with montelukast compared to placebo may result in a reduction of pruritus, but the evidence is very uncertain (two studies, 87 participants): SMD -1.40, 95% CI -1.87 to -0.92; certainty of evidence: very low. Treatment with fish-oil/omega-3 fatty acids compared to placebo may result in a large reduction of pruritus (four studies, 160 observations): SMD -1.60, 95% CI -1.97 to -1.22; certainty of evidence: low. Treatment with cromolyn sodium compared to placebo may result in a reduction of pruritus, but the evidence is very uncertain (VAS 0 to 10 cm, MD -3.27, 95% CI -5.91 to -0.63; two RCTs, N = 100, certainty of evidence: very low). Treatment with topical capsaicin compared with placebo may result in a large reduction of pruritus (two studies; 112 participants): SMD -1.06, 95% CI -1.55 to -0.57; certainty of evidence: low. Ondansetron, zinc sulphate and several other treatments may not reduce pruritus in participants suffering from UP. In participants with cholestatic pruritus (CP), treatment with rifampicin compared to placebo may reduce pruritus, but the evidence is very uncertain (VAS: 0 to 100, MD -42.00, 95% CI -87.31 to 3.31; two RCTs, N = 42, certainty of evidence: very low). Treatment with flumecinol compared to placebo may reduce pruritus, but the evidence is very uncertain (RR > 1 favours treatment group; RR 2.32, 95% CI 0.54 to 10.1; two RCTs, N = 69, certainty of evidence: very low). Treatment with the opioid antagonist naltrexone compared to placebo may reduce pruritus (VAS: 0 to 10 cm, MD -2.42, 95% CI -3.90 to -0.94; two RCTs, N = 52, certainty of evidence: low). However, effects in participants with UP were inconclusive (percentage of difference -12.30%, 95% CI -25.82% to 1.22%, one RCT, N = 32). In palliative care participants with pruritus of a different nature, the treatment with the drug paroxetine (one study), a selective serotonin reuptake inhibitor, compared to placebo may reduce pruritus slightly by 0.78 (numerical analogue scale from 0 to 10 points; 95% CI -1.19 to -0.37; one RCT, N = 48, certainty of evidence: low). Most adverse events were mild or moderate. Two interventions showed multiple major adverse events (naltrexone and nalfurafine). AUTHORS CONCLUSIONS: Different interventions (GABA-analogues, kappa-opioid receptor agonists, cromolyn sodium, montelukast, fish-oil/omega-3 fatty acids and topical capsaicin compared to placebo) were effective for uraemic pruritus. GABA-analogues had the largest effect on pruritus. Rifampin, naltrexone and flumecinol tended to be effective for cholestatic pruritus. However, therapies for patients with malignancies are still lacking. Due to the small sample sizes in most meta-analyses and the heterogeneous methodological quality of the included trials, the results should be interpreted cautiously in terms of generalisability.

Interventions for preventing back pain among office workers - a systematic review and network meta-analysis.

OBJECTIVE: Back pain is common in the working population. This systematic review with network meta-analysis (NMA) aimed to compare the effects of interventions for preventing back pain among office workers. METHODS: We searched eight databases and additional sources up to March 2021. We included randomized controlled trials (RCT) and cluster RCT focusing on office workers, comparing work-related interventions aimed at preventing back pain (defined as pain in any part of the spine) to a control condition and assessing back pain and/or work absence. Further outcomes considered were adverse events and participants' satisfaction. We performed both frequentist and component NMA. Risk of bias (RoB) was evaluated using RoB 2 and certainty of the evidence (CoE) was assessed using GRADE. RESULTS: We screened 9809 records and included 24 studies with a total of 7080 participants. RoB was assessed as "some concerns" or "high" for all studies and outcomes. Included studies investigated multicomponent interventions, ergonomics, physical activity, education, behavioral interventions and no/minimal interventions. Effects were mostly not statistically significant and based on low/very low CoE. Physical activity probably reduces days of work absence slightly [mean difference (MD) -1.10, 95% confidence interval (CI) -2.07- -0.13], and combining physical activity and ergonomics may reduce back pain intensity (standardized MD -0.41, 95% CI -0.80- -0.02) when compared to no/minimal intervention. A large proportion of participants were satisfied with the interventions, adverse events were rarely assessed. CONCLUSIONS: We observed mostly minor effects of interventions on back pain and work absence among office workers. The practical relevance of these effects is questionable.

Valve-in-Valve Transcatheter Aortic Valve Replacement Versus Redo Surgical Aortic Valve Replacement for Failed Surgical Aortic Bioprostheses: A Systematic Review and Meta-Analysis.

Background In the absence of randomized controlled trials, reports from nonrandomized studies comparing valve-in-valve implantation (ViV) to redo surgical aortic valve replacement (rAVR) have shown inconsistent results. Methods and Results PubMed/MEDLINE, Google Scholar, and CENTRAL (Cochrane Central Register of Controlled Trials) were searched through December 2021. Meta-Analysis of Observational Studies in Epidemiology guidelines were followed. The protocol was registered at the International Prospective Register of Systematic Reviews. Random effects models were applied. The primary outcomes of interest were short-term and midterm mortality. Secondary outcomes included stroke, myocardial infarction, acute renal failure, and permanent pacemaker implantation, as well as prosthetic aortic valve regurgitation, mean transvalvular gradient, and severe prosthesis-patient mismatch. Of 8881 patients included in 15 studies, 4458 (50.2%) underwent ViV and 4423 (49.8%) rAVR. Short-term mortality was 2.8% in patients undergoing ViV compared with 5.0% in patients undergoing rAVR (risk ratio [RR] 0.55 [95% CI, 0.34-0.91], P=0.02). Midterm mortality did not differ in patients undergoing ViV compared with patients undergoing rAVR (hazard ratio, 1.27 [95% CI, 0.72-2.25]). The rate of acute kidney failure was lower following ViV, (RR, 0.54 [95% CI, 0.33-0.88], P=0.02), whereas prosthetic aortic valve regurgitation (RR, 4.18 [95% CI, 1.88-9.3], P=0.003) as well as severe patient-prothesis mismatch (RR, 3.12 [95% CI, 2.35-4.1], P<0.001) occurred more frequently. The mean transvalvular gradient was higher following ViV (standard mean difference, 0.44 [95% CI, 0.15-0.72], P=0.008). There were no significant differences between groups with respect to stroke (P=0.26), myocardial infarction (P=0.93), or pacemaker implantation (P=0.21). Conclusions Results of this meta-analysis demonstrate better short-term mortality after ViV compared with rAVR. Midterm mortality was similar between groups. Given the likely selection bias in these individual reports, an adequately powered multicenter randomized clinical trial with sufficiently long follow-up in patients with low-to-intermediate surgical risk is warranted. Registration URL: crd.york.ac.uk/prospero/. Unique identifier: CRD42021228752.

Open Surgical Thrombendarterectomy Versus Endovascular Treatment in Occlusive Processes of the Femoral Artery Bifurcation.

BACKGROUND: The standard treatment of occlusive processes of the femoral artery bifurcation is thrombendarterectomy (TEA). Endovascular techniques (ENDO) have recently been put forward as a potential alternative. It is unclear so far which modality yields better outcomes with respect to long-term revascularization and periprocedural complications. Method: Multiple databases were systematically searched for pertinent publications (publication date November 1965 to February 2022). From the included studies, individual patient data (IPD) were requested. Aggregate data (AD) were used when no IPD were available. Primary and secondary patency (PP and SP), perioperative morbidity/mortality, and further endpoints were determined separately for TEA and ENDO and compared with each other. AD for each modality were summarized in meta-analyses. Time-to-event analyses and comparative meta-analyses with PP as primary endpoint were carried out using IPD. Results: 42 studies (3 IPD, 39 AD; 27 TEA, 12 ENDO, 3 comparisons of TEA versus ENDO) were included. In the combined meta-analysis of IPD and AD, PP for TEA was 97% at 6 months and 92% at 12 months, while PP for ENDO was 84% at 6 months and 85% at 12 months. The differences were not statistically significant. The comparative meta-analysis regarding PP did not reveal any significant differences either (TEA versus ENDO: HR 0.30 [0.06; 1.48]). SP at 12 months was 97% (TEA) and 93% (ENDO). The periprocedural morbidity was 16% for TEA and 9% for ENDO. Conclusion: In light of a higher PP, even without formal statistical proof of superiority, TEA can still be considered the standard treatment for occlusive processes of the femoral artery bifurcation. METHODS: Multiple databases were systematically searched for pertinent publications (publication date November 1965 to February 2022). From the included studies, individual patient data (IPD) were requested. Aggregate data (AD) were used when no IPD were available. Primary and secondary patency (PP and SP), perioperative morbidity/mortality, and further endpoints were determined separately for TEA and ENDO and compared with each other. AD for each modality were summarized in meta-analyses. Time-to-event analyses and comparative meta-analyses with PP as primary endpoint were carried out using IPD. RESULTS: 42 studies (3 IPD, 39 AD; 27 TEA, 12 ENDO, 3 comparisons of TEA versus ENDO) were included. In the combined meta-analysis of IPD and AD, PP for TEA was 97% at 6 months and 92% at 12 months, while PP for ENDO was 84% at 6 months and 85% at 12 months. The differences were not statistically significant. The comparative meta-analysis regarding PP did not reveal any significant differences either (TEA versus ENDO: HR 0.30 [0.06; 1.48]). SP at 12 months was 97% (TEA) and 93% (ENDO). The periprocedural morbidity was 16% for TEA and 9% for ENDO. CONCLUSION: In light of a higher PP, even without formal statistical proof of superiority, TEA can still be considered the standard treatment for occlusive processes of the femoral artery bifurcation.

Living systematic review and meta-analysis of the prostate MRI diagnostic test with Prostate Imaging Reporting and Data System (PI-RADS) assessment for the detection of prostate cancer: study protocol.

INTRODUCTION: The Prostate Imaging Reporting and Data System (PI-RADS) standardises reporting of prostate MRI for the detection of clinically significant prostate cancer. We provide the protocol of a planned living systematic review and meta-analysis for (1) diagnostic accuracy (sensitivity and specificity), (2) cancer detection rates of assessment categories and (3) inter-reader agreement. METHODS AND ANALYSIS: Retrospective and prospective studies reporting on at least one of the outcomes of interest are included. Each step that requires literature evaluation and data extraction is performed by two independent reviewers. Since PI-RADS is intended as a living document itself, a 12-month update cycle of the systematic review and meta-analysis is planned.This protocol is in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses-Protocols statement. The search strategies including databases, study eligibility criteria, index and reference test definitions, outcome definitions and data analysis processes are detailed. A full list of extracted data items is provided.Summary estimates of sensitivity and specificity (for PI-RADS ≥3 and PI-RADS ≥4 considered positive) are derived with bivariate binomial models. Summary estimates of cancer detection rates are calculated with random intercept logistic regression models for single proportions. Summary estimates of inter-reader agreement are derived with random effects models. ETHICS AND DISSEMINATION: No original patient data are collected, ethical review board approval, therefore, is not necessary. Results are published in peer-reviewed, open-access scientific journals. We make the collected data accessible as supplemental material to guarantee transparency of results. PROSPERO REGISTRATION NUMBER: CRD42022343931.

Hyperthyroidism and clinical depression: a systematic review and meta-analysis.

Hyperthyroidism and clinical depression are common, and there is preliminary evidence of substantial comorbidity. The extent of the association in the general population, however, has not yet been estimated meta-analytically. Therefore we conducted this systematic review and meta-analysis (registered in PROSPERO: CRD42020164791). Until May 2020, Medline (via PubMed), PsycINFO, and Embase databases were systematically searched for studies on the association of hyperthyroidism and clinical depression, without language or date restrictions. Two reviewers independently selected epidemiological studies providing laboratory or ICD-based diagnoses of hyperthyroidism and diagnoses of depression according to operationalized criteria (e.g. DSM) or to cut-offs in established rating scales. All data, including study quality based on the Newcastle-Ottawa Scale, were independently extracted by two authors. Odds ratios for the association of clinical depression and hyperthyroidism were calculated in a DerSimonian-Laird random-effects meta-analysis. Out of 3372 papers screened we selected 15 studies on 239 608 subjects, with 61% women and a mean age of 50. Relative to euthyroid individuals, patients with hyperthyroidism had a higher chance of being diagnosed with clinical depression: OR 1.67 ([95% CI: 1.49; 1.87], I2: 6%; prediction interval: 1.40 to 1.99), a result supported in a number of sensitivity and subgroup analyses. The OR was slightly less pronounced for subclinical as opposed to overt hyperthyroidism (1.36 [1.06; 1.74] vs. 1.70 [1.49; 1.93]). This comorbidity calls for clinical awareness and its reasons need investigation and may include neurobiological mechanisms, common genetic vulnerability and a generally heightened risk for clinical depression in patients with chronic somatic disorders.

Dose effects of tricyclic antidepressants in the treatment of acute depression - A systematic review and meta-analysis of randomized trials.

BACKGROUND: Tricyclic antidepressants (TCA) continue to be an important group of drugs, but it is unclear whether a dose-response relationship is supported by high-level evidence. METHODS: Systematic review in the Cochrane Collaboration's Central Register of Controlled Trials (CENTRAL) of studies randomizing patients to at least two doses of one TCA, complemented by searches in Medline, Embase, and PsycInfo. In multilevel regression, we calculated the standardized mean difference (SMD) in antidepressant efficacy per mg TCA dose increase, and we analyzed drop-outs due to adverse events. Finally, we computed random effects meta-analyses of all dose comparisons investigated in a minimum of two studies. RESULTS: Out of 5365 studies screened, we included 15 randomized trials on 24 comparisons of 14 different dose contrasts. We found a statistically non-significant positive effect of increasing the dose: 0.34 SMD with 100 mg/d dose increase ([-0.03; 0.70] p = 0.073). While several comparisons showed no clear signal of a dose gradient, 300 mg of imipramine/desipramine is statistically significantly superior to 150 mg (SMD: 0.80 [0.28; 1.33], p = 0.003, I2: 0%). Drop-outs increased with higher doses, albeit not statistically significantly: Odds ratio (OR) of 1.44 with 100 mg dose increase [0.54; 3.86]. Overall, risk of bias was high. LIMITATIONS: Limited number of studies with mainly high risk of bias. CONCLUSIONS: So far, data on a dose-response relationship in TCAs from direct dose comparisons are inconclusive. Clinically, escalation to high doses may be justified if side effects are bearable.

An Empirical Evaluation of the Impact Scenario of Pooling Bodies of Evidence from Randomized Controlled Trials and Cohort Studies in Nutrition Research.

Only very few Cochrane nutrition reviews include cohort studies (CSs), but most evidence in nutrition research comes from CSs. We aimed to pool bodies of evidence (BoE) from randomized controlled trials (RCTs) derived from Cochrane reviews with matched BoE from CSs. The Cochrane Database of Systematic Reviews and MEDLINE were searched for systematic reviews (SRs) of RCTs and SRs of CSs. BoE from RCTs were pooled together with BoE from CSs using random-effects and common-effect models. Heterogeneity, 95% prediction intervals, contributed weight of BoE from RCTs to the pooled estimate, and whether integration of BoE from CSs modified the conclusion from BoE of RCTs were evaluated. Overall, 80 diet-disease outcome pairs based on 773 RCTs and 720 CSs were pooled. By pooling BoE from RCTs and CSs with a random-effects model, for 45 (56%) out of 80 diet-disease associations the 95% CI excluded no effect and showed mainly a reduced risk/inverse association. By pooling BoE from RCTs and CSs, median I2 = 46% and the median contributed weight of RCTs to the pooled estimates was 34%. The direction of effect between BoE from RCTs and pooled effect estimates was rarely opposite (n = 17; 21%). The integration of BoE from CSs modified the result (by examining the 95% CI) from BoE of RCTs in 35 (44%) of the 80 diet-disease associations. Our pooling scenario showed that the integration of BoE from CSs modified the conclusion from BoE of RCTs in nearly 50% of the associations, although the direction of effect was mainly concordant between BoE of RCTs and pooled estimates. Our findings provide insights for the potential impact of pooling both BoE in Cochrane nutrition reviews. CSs should be considered for inclusion in future Cochrane nutrition reviews, and we recommend analyzing RCTs and CSs in separate meta-analyses, or, if combined together, with a subgroup analysis.

Bayesian models for aggregate and individual patient data component network meta-analysis.

Network meta-analysis can synthesize evidence from studies comparing multiple treatments for the same disease. Sometimes the treatments of a network are complex interventions, comprising several independent components in different combinations. A component network meta-analysis (CNMA) can be used to analyze such data and can in principle disentangle the individual effect of each component. However, components may interact with each other, either synergistically or antagonistically. Deciding which interactions, if any, to include in a CNMA model may be difficult, especially for large networks with many components. In this article, we present two Bayesian CNMA models that can be used to identify prominent interactions between components. Our models utilize Bayesian variable selection methods, namely the stochastic search variable selection and the Bayesian LASSO, and can benefit from the inclusion of prior information about important interactions. Moreover, we extend these models to combine data from studies providing aggregate information and studies providing individual patient data (IPD). We illustrate our models in practice using three real datasets, from studies in panic disorder, depression, and multiple myeloma. Finally, we describe methods for developing web-applications that can utilize results from an IPD-CNMA, to allow for personalized estimates of relative treatment effects given a patient's characteristics.

Answering complex hierarchy questions in network meta-analysis.

BACKGROUND: Network meta-analysis estimates all relative effects between competing treatments and can produce a treatment hierarchy from the most to the least desirable option according to a health outcome. While about half of the published network meta-analyses present such a hierarchy, it is rarely the case that it is related to a clinically relevant decision question. METHODS: We first define treatment hierarchy and treatment ranking in a network meta-analysis and suggest a simulation method to estimate the probability of each possible hierarchy to occur. We then propose a stepwise approach to express clinically relevant decision questions as hierarchy questions and quantify the uncertainty of the criteria that constitute them. The steps of the approach are summarized as follows: a) a question of clinical relevance is defined, b) the hierarchies that satisfy the defined question are collected and c) the frequencies of the respective hierarchies are added; the resulted sum expresses the certainty of the defined set of criteria to hold. We then show how the frequencies of all possible hierarchies relate to common ranking metrics. RESULTS: We exemplify the method and its implementation using two networks. The first is a network of four treatments for chronic obstructive pulmonary disease where the most probable hierarchy has a frequency of 28%. The second is a network of 18 antidepressants, among which Vortioxetine, Bupropion and Escitalopram occupy the first three ranks with frequency 19%. CONCLUSIONS: The developed method offers a generalised approach of producing treatment hierarchies in network meta-analysis, which moves towards attaching treatment ranking to a clear decision question, relevant to all or a subset of competing treatments.

Combining Antidepressants vs Antidepressant Monotherapy for Treatment of Patients With Acute Depression: A Systematic Review and Meta-analysis.

IMPORTANCE: Combining antidepressants is frequently done in the treatment of acute depression, but studies have yielded conflicting results. OBJECTIVE: To conduct a systematic review and meta-analysis assessing efficacy and tolerability of combination therapy. Combinations using presynaptic α2-autoreceptor antagonists or bupropion were investigated separately. DATA SOURCES: MEDLINE, Embase, PsycINFO, and the Cochrane Central Register of Controlled Trials were systematically searched from each database inception through January 2020. STUDY SELECTION: Randomized clinical trials (RCTs) comparing combinations of antidepressants with antidepressant monotherapy in adult patients with acute depression were included. DATA EXTRACTION AND SYNTHESIS: Following guidelines from Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) and recommendations from the Cochrane Handbook, 2 reviewers independently performed a literature search, study selection, data extraction, and evaluation of risk of bias. Data were pooled in random-effects analyses. MAIN OUTCOMES AND MEASURES: Primary outcome was efficacy measured as standardized mean difference (SMD); secondary outcomes were response, remission, change from baseline in rating scale scores, number of dropouts, and number of dropouts due to adverse events. RESULTS: Thirty-nine RCTs including 6751 patients were eligible. Combination treatment was statistically significantly associated with superior treatment outcomes relative to monotherapy (SMD = 0.31; 95% CI, 0.19-0.44). Combining a reuptake inhibitor with an antagonist of presynaptic α2-autoreceptors was superior to other combinations (SMD = 0.37; 95% CI, 0.19-0.55). Bupropion combinations were not superior to monotherapy (SMD = 0.10; 95% CI, -0.07 to 0.27). Numbers of dropouts and dropouts due to adverse events did not differ between treatments. Studies were heterogeneous, and there was indication of publication bias (Egger test result was positive; P = .007, df = 36), but results remained robust across prespecified secondary outcomes and sensitivity and subgroup analyses, including analyses restricted to studies with low risk of bias. CONCLUSIONS AND RELEVANCE: In this meta-analysis of RCTs comparing combinations of antidepressants with antidepressant monotherapy, combining antidepressants was associated with superior treatment outcomes but not with more patients dropping out of treatment. Combinations using an antagonist of presynaptic α2-autoreceptors may be preferable and may be applied as a first-line treatment in severe cases of depression and for patients considered nonresponders.

Effects of nutritional intervention strategies in the primary prevention of overweight and obesity in school settings: systematic review and network meta-analysis.

Objective: To examine the effects of different nutritional intervention strategies in the school setting on anthropometric and quality of diet outcomes by comparing and ranking outcomes in a network meta-analysis. Design: Systematic review and network meta-analysis. Data sources: PubMed, Cochrane Central Register of Controlled Trials (CENTRAL), Web of Science, Education Resources Information Centre (ERIC), PsycInfo, CAB Abstracts, Campbell Library, Evidence for Policy and Practice Information and Co-ordinating Centre (EPPI-Centre) BiblioMap, Australian Education Index, Joanna Briggs Institute Evidence-Based Practice (JBI EBP) database, Practice-based Evidence in Nutrition (PEN) database, ClinicalTrials.gov, Current Controlled Trials, and World Health Organization International Clinical Trials Registry Platform. Eligibility criteria for selecting studies: A systematic literature search was performed from inception to 2 May 2022. Cluster randomised controlled trials meeting these study criteria were included: generally healthy school students aged 4-18 years; intervention with ≥1 nutritional components in a school setting; and studies that assessed anthropometric measures (eg, body mass index, body fat) or measures related to the quality of diet (eg, intake of fruit and vegetables), or both. Random effects pairwise meta-analyses and network meta-analyses were performed with a frequentist approach. P scores, a frequentist analogue to surface under the cumulative ranking curve, ranging from 0 to 1 (indicating worst and best ranked interventions, respectively) were calculated. Risk of bias was assessed with Cochrane's RoB 2 tool. The Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework was used to rate the certainty of evidence. Results: 51 cluster randomised controlled trials involving 75 954 participants and seven intervention nodes were included. Inconsistency could not be assessed (except for intake of fruit and vegetables) because the network meta-analyses were based mainly on star shaped networks with no direct evidence for specific pairs of nutritional interventions. Overall, little or no evidence was found to support a difference in body mass index, body weight, body fat, or waist circumference and moderate improvements in intake of fruit and vegetables with nutritional interventions in a school setting. Low to moderate certainty of evidence further suggested that multicomponent nutritional interventions likely reduced the prevalence (odds ratio 0.66, 95% confidence interval 0.55 to 0.80) and incidence (0.67, 0.47 to 0.96) of overweight compared with a control group. Based on low certainty of evidence, nutrition education and multicomponent interventions may be more effective than a control group (ie, usual practice) for increasing intake of fruit and vegetables. Multicomponent nutritional interventions were ranked the most effective for reducing body mass index (P score 0.76) and intake of fat (0.82). Nutrition education was ranked as best for body mass index z score (0.99), intake of fruit and vegetables (0.82), intake of fruit (0.92), and intake of vegetables (0.88). Conclusions: The findings suggest that nutritional interventions in school settings may improve anthropometric and quality of diet measures, potentially contributing to the prevention of overweight and obesity in childhood and adolescence. The findings should be interpreted with caution because the certainty of evidence was often rated as low. The results of the network meta-analysis could be used by policy makers in developing and implementing effective, evidence based nutritional intervention strategies in the school setting. Systematic review registration: PROSPERO CRD42020220451.

Subsequent full publication of qualitative studies presented at United Kingdom Royal College of Nursing Research Conference 2015 and 2016: A follow-up study.

A considerable proportion of quantitative research remains unpublished once completed. Little research has documented non-dissemination and dissemination bias in qualitative research. This study aimed to generate evidence on the extent of non-dissemination in qualitative research. We followed a cohort of qualitative studies presented as conference abstracts to ascertain their subsequent publication status. We searched for subsequent full publication in MEDLINE, in the Cumulative Index to Nursing & Allied Health Literature and in Google Scholar. We matched abstracts to subsequent publications according to authors, method of data collection and phenomenon of interest. Fisher's exact test was calculated to examine associations between study characteristics and publication. Factors potentially associated with time to publication were evaluated with Cox regression analysis. For 91 of 270 included abstracts (33.70%; 95% CI 28.09%-39.68%), no full publication was identified. Factors that were found to be associated with subsequent full publication were oral presentation (OR 4.62; 95% CI 2.43-8.94) and university affiliation (OR 1.96; 95% CI 1.05-3.66). Compared to oral presentations, studies presented as posters took longer time to reach full publication (hazard ratio 0.35, 95% CI 0.21-0.58). This study shows that it was not possible to retrieve a full publication for over one-third of abstracts. Our findings suggest that where this non-dissemination is systematic, it may lead to distortions of the qualitative evidence-base for decision-making through dissemination bias. Our findings are congruent with those of other studies. Further research might investigate non-dissemination of qualitative studies in other disciplines to consolidate our findings.

Reevaluation of statistically significant meta-analyses in advanced cancer patients using the Hartung-Knapp method and prediction intervals-A methodological study.

Using the Hartung-Knapp method and 95% prediction intervals (PIs) in random-effects meta-analyses is recommended by experts but rarely applied. Therefore, we aimed to reevaluate statistically significant meta-analyses using the Hartung-Knapp method and 95% PIs. In this methodological study, three databases were searched from January 2010 to July 2019. We included systematic reviews reporting a statistically significant meta-analysis of at least four randomized controlled trials in advanced cancer patients using either a fixed-effect or random-effects model. We investigated the impact of switching from fixed-effect to random-effects meta-analysis and of using the recommended Hartung-Knapp method in random-effects meta-analyses. Furthermore, we calculated 95% PIs for all included meta-analyses. We identified 6234 hits, of which 261 statistically significant meta-analyses were included. Our recalculations of these 261 meta-analyses produced statistically significant results in 132 of 138 fixed-effect and 114 of 123 random-effects meta-analyses. When switching to a random-effects model, 19 of 132 fixed-effect meta-analyses (14.4%) were no longer statistically significant. Using the Hartung-Knapp method in random-effects meta-analyses resulted in 34 of 114 nonsignificant meta-analyses (29.8%). In the full sample (N = 261), the null effect was included by the 95% PI in 195 (74.7%) and the opposite effect (e.g., hazard ratio 0.5, opposite effect 2) in 98 meta-analyses (37.5%). Using the Hartung-Knapp method and PIs substantially influenced the interpretation of many published, statistically significant meta-analyses. We strongly encourage researchers to check if using the Hartung-Knapp method and reporting 95% PIs is appropriate in random-effects meta-analyses.

Meta-Analysis of Proportions.

The meta-analysis of single proportions has become a popular application over the last two decades. Especially, systematic reviews of prevalence studies are conducted in various fields of science, including medicine, ecology, psychology, or social sciences. In this chapter, we illustrate meta-analysis methods to pool single proportions and to compare proportions from two groups. We introduce classic approaches based on the inverse variance method as well as generalized linear mixed models taking the binary structure of the data into account. The most common transformations of proportions and their back-transformations are described both for individual studies and in the meta-analysis setting.

Prevalence of evidence of inconsistency and its association with network structural characteristics in 201 published networks of interventions.

BACKGROUND: Network meta-analysis (NMA) has attracted growing interest in evidence-based medicine. Consistency between different sources of evidence is fundamental to the reliability of the NMA results. The purpose of the present study was to estimate the prevalence of evidence of inconsistency and describe its association with different NMA characteristics. METHODS: We updated our collection of NMAs with articles published up to July 2018. We included networks with randomised clinical trials, at least four treatment nodes, at least one closed loop, a dichotomous primary outcome, and available arm-level data. We assessed consistency using the design-by-treatment interaction (DBT) model and testing all the inconsistency parameters globally through the Wald-type chi-squared test statistic. We estimated the prevalence of evidence of inconsistency and its association with different network characteristics (e.g., number of studies, interventions, intervention comparisons, loops). We evaluated the influence of the network characteristics on the DBT p-value via a multivariable regression analysis and the estimated Pearson correlation coefficients. We also evaluated heterogeneity in NMA (consistency) and DBT (inconsistency) random-effects models. RESULTS: We included 201 published NMAs. The p-value of the design-by-treatment interaction (DBT) model was lower than 0.05 in 14% of the networks and lower than 0.10 in 20% of the networks. Networks including many studies and comparing few interventions were more likely to have small DBT p-values (less than 0.10), which is probably because they yielded more precise estimates and power to detect differences between designs was higher. In the presence of inconsistency (DBT p-value lower than 0.10), the consistency model displayed higher heterogeneity than the DBT model. CONCLUSIONS: Our findings show that inconsistency was more frequent than what would be expected by chance, suggesting that researchers should devote more resources to exploring how to mitigate inconsistency. The results of this study highlight the need to develop strategies to detect inconsistency (because of the relatively high prevalence of evidence of inconsistency in published networks), and particularly in cases where the existing tests have low power.