61 research outputs found
A primer on network meta-analysis with emphasis on mental health
No full textA quantitative synthesis of evidence via standard pair-wise meta-analysis lies on the top of the hierarchy for evaluating the relative effectiveness or safety between two interventions. In most healthcare problems, however, there is a plethora of competing interventions. Network meta-analysis allows to rank competing interventions and evaluate their relative effectiveness even if they have not been compared in an individual trial. The aim of this paper is to explain and discuss the main features of this statistical technique
An approach for modelling multiple correlated outcomes in a network of interventions using odds ratios.
No full textA multivariate meta-analysis of two or more correlated outcomes is expected to improve precision compared with a series of independent, univariate meta-analyses especially when there are studies reporting some but not all outcomes. Multivariate meta-analysis requires estimates of the within-study correlations, which are seldom available. Existing methods for analysing multiple outcomes simultaneously are limited to pairwise treatment comparisons. We propose a model for a joint, simultaneous synthesis of multiple dichotomous outcomes in a network of interventions and introduce a simple way to elicit expert opinion for the within-study correlations by utilizing a set of conditional probability parameters. We implement our multiple-outcomes network meta-analysis model within a Bayesian framework, which allows incorporation of expert information. As an example, we analyse two correlated dichotomous outcomes, response to the treatment and dropout rate, in a network of pharmacological interventions for acute mania. The produced estimates have narrower confidence intervals compared with the simple network meta-analysis. We conclude that the proposed model and the suggested prior elicitation method for correlations constitute a useful framework for performing network meta-analysis for multiple outcomes
Joint synthesis of multiple correlated outcomes in networks of interventions.
No full textMultiple outcomes multivariate meta-analysis (MOMA) is gaining in popularity as a tool for jointly synthesizing evidence coming from studies that report effect estimates for multiple correlated outcomes. Models for MOMA are available for the case of the pairwise meta-analysis of two treatments for multiple outcomes. Network meta-analysis (NMA) can be used for handling studies that compare more than two treatments; however, there is currently little guidance on how to perform an MOMA for the case of a network of interventions with multiple outcomes. The aim of this paper is to address this issue by proposing two models for synthesizing evidence from multi-arm studies reporting on multiple correlated outcomes for networks of competing treatments. Our models can handle continuous, binary, time-to-event or mixed outcomes, with or without availability of within-study correlations. They are set in a Bayesian framework to allow flexibility in fitting and assigning prior distributions to the parameters of interest while fully accounting for parameter uncertainty. As an illustrative example, we use a network of interventions for acute mania, which contains multi-arm studies reporting on two correlated binary outcomes: response rate and dropout rate. Both multiple-outcomes NMA models produce narrower confidence intervals compared with independent, univariate network meta-analyses for each outcome and have an impact on the relative ranking of the treatments
Percutaneous Coronary Interventions for the Treatment of Stenoses in Small Coronary Arteries: A Network Meta-Analysis.
No full textOBJECTIVES
This study evaluated the most appropriate percutaneous coronary intervention (PCI) for the treatment of stenoses in small coronary arteries.
BACKGROUND
PCI in small coronary arteries is associated with an increased risk of lesion failure and restenosis.
METHODS
Randomized trials comparing different PCI strategies were identified through a broad search of published reports. Primary angiographic outcome was %DS (%DS). A pairwise meta-analysis was performed by using random effects model, followed by a network meta-analysis synthesizing direct and indirect evidence.
RESULTS
Overall, 19 trials were eligible, which included 5,072 patients comprising a network without closed loops among 5 identified interventions (early generation sirolimus-eluting stents [SES], paclitaxel-eluting stents [PES], drug-coated balloons [DCB], bare-metal stents [BMS], and balloon angioplasty [BA]). No dedicated trial was identified evaluating new generation drug-eluting stents. Early generation SES yielded the best angiographic results according to %DS. For %DS, SES was ranked as the most effective treatment, followed by PES (standardized mean differences [SMD]: -0.44; 95% confidence interval [CI]: -0.92 to 0.05 vs. SES) and DCB (SMD: -0.89; 95% CI: -1.53 to -0.25 vs. SES). In terms of absolute differences, SES yielded a reduction of 18% in diameter stenosis compared to DCB. SES significantly reduced the risk of target-lesion revascularization compared to PES (odds ratio [OR]: 0.39; 95% CI: 0.16 to 0.93), DCB (OR: 0.34; 95% CI: 0.10 to 0.97), BMS (OR: 0.21; 95% CI: 0.13 to 0.36), and BA (OR: 0.16; 95% CI: 0.09 to 0.29).
CONCLUSIONS
Early generation SES yielded the most favorable angiographic and clinical outcomes for the treatment of stenoses in small coronary arteries. New generation DES need to be evaluated against this standard in future randomized trials
Effect of statins and non-statin LDL-lowering medications on cardiovascular outcomes in secondary prevention: A meta-analysis of randomized trials
No full textAIMS:
Current evidence on dyslipidaemia management has expanded to novel treatments and very low achieved levels of low-density lipoprotein cholesterol (LDL-C). We sought to compare the clinical impact of more-intensive vs. less-intensive LDL-C lowering by means of statins and currently recommended non-statin medications in secondary prevention.
METHODS AND RESULTS:
We searched Medline, EMBASE, and Cochrane databases for randomized controlled trials of statins, ezetimibe, proprotein convertase subtilisin-kexin type 9 (PCSK9) inhibitors, or bile acid sequestrants with >500 patients followed for ≥1 year. We employed random-effects models using risk ratios (RRs) with 95% confidence intervals (CIs) to compare outcomes. We included 19 trials (15 of statins, 3 of PCSK9 inhibitors, and 1 of ezetimibe) with 152 507 patients randomly assigned to more-intensive (n = 76 678) or less-intensive treatment (n = 75 829). More-intensive treatment was associated with 19% relative risk reduction for the primary outcome, major vascular events (MVEs; RR 0.81, 95% CI 0.77-0.86). Risk reduction was greater across higher baseline levels and greater achieved reductions of LDL-C. The clinical benefit was significant across varying types of more-intensive treatment and was consistent for statins (RR 0.81, 95% CI 0.76-0.86) and non-statin agents (PCSK9 inhibitors and ezetimibe; RR 0.85, 95% CI 0.77-0.94) as active (more-intensive) intervention (P-interaction = 0.38). Each 1.0 mmol/L reduction in LDL-C was associated with 19% relative decrease in MVE. Death, cardiovascular death, myocardial infarction, stroke, and coronary revascularization also favoured more-intensive treatment.
CONCLUSION:
Reduction of MVE is proportional to the magnitude of LDL-C lowering across a broad spectrum of on-treatment levels in secondary prevention. Statin intensification and add-on treatment with PCSK9 inhibitors or ezetimibe are associated with significant reduction of cardiovascular morbidity in this very high-risk population
Allowing for uncertainty due to missing continuous outcome data in pairwise and network meta-analysis
No full textMissing outcome data are commonly encountered in randomized controlled trials and hence may need to be addressed in a meta-analysis of multiple trials. A common and simple approach to deal with missing data is to restrict analysis to individuals for whom the outcome was obtained (complete case analysis). However, estimated treatment effects from complete case analyses are potentially biased if informative missing data are ignored. We develop methods for estimating meta-analytic summary treatment effects for continuous outcomes in the presence of missing data for some of the individuals within the trials. We build on a method previously developed for binary outcomes, which quantifies the degree of departure from a missing at random assumption via the informative missingness odds ratio. Our new model quantifies the degree of departure from missing at random using either an informative missingness difference of means or an informative missingness ratio of means, both of which relate the mean value of the missing outcome data to that of the observed data. We propose estimating the treatment effects, adjusted for informative missingness, and their standard errors by a Taylor series approximation and by a Monte Carlo method. We apply the methodology to examples of both pairwise and network meta-analysis with multi-arm trials
Variation in the surgical management of complicated diverticulitis: a cross-sectional study of European surgeons
No full textIntroduction: There are many options for the surgical management of complicated diverticulitis, and standards vary widely despite international practice recommendations. We conducted a survey to capture the variation in practice across Europe.
Methods: An online questionnaire was distributed to fellow and surgeon members of the European Association of Endoscopic Surgery (EAES) via email using the Opinio survey platform. Participants shared their demographic details. We asked members to rank the most likely intervention for patients with both stable and unstable Hinchey Class III, as well as Hinchey Class IV diverticulitis based on practice standards in their country. We used descriptive statistics, including counts and percentages, to characterize survey results. We created a heatmap to visualize the percentage of votes received for each intervention.
Results: We received 233 responses from surgeons and fellows across Europe from various countries, including Italy (35.6%), Greece (11.2%), and the United Kingdom (9.9%). Most members (79.4%) self-reported having expertise in colorectal surgery. For patients with stable Hinchey III diverticulitis, surgeons offered Hartmann's resection (HR) (41.6%), primary resection and anastomosis (PRA) (18.5%), laparoscopic peritonea lavage (LPL) prior to HR (16.9%), or LPL prior to PRA (15.5%), or LPL only (8.6%). In total, 31.4% of respondents offered LPL prior to sigmoid resection (HR + PRA). For patients with unstable Hinchey III diverticulitis, respondents offered HR (73.9%), PRA (3.85%), LPL only (6.84%), or LPL followed by sigmoid resection (15.4%). For patients with stable Hinchey IV diverticulitis, surgeons offered HR (71.7%), PRA (4.7%), LPL only (1.3%), or LPL then sigmoid resection (22.3%). Finally, for patients with unstable Hinchey IV diverticulitis, surgeons offered HR (83.1%), PRA (1.3%), LPL only (3.5%), or LPL followed by sigmoid resection (12.1%).
Conclusion: Significant variation exists in the surgical management of complicated diverticulitis across Europe. Efforts must be made to increase the awareness and uptake of surgical guideline recommendations in clinical practice
A review of verbal and non-verbal human–robot interactive communication
No full textAbstractIn this paper, an overview of human–robot interactive communication is presented, covering verbal as well as non-verbal aspects. Following a historical introduction, and motivation towards fluid human–robot communication, ten desiderata are proposed, which provide an organizational axis both of recent as well as of future research on human–robot communication. Then, the ten desiderata are examined in detail, culminating in a unifying discussion, and a forward-looking conclusion
Recent meta-analyses neglect previous systematic reviews and meta-analyses about the same topic: a systematic examination
No full textAs the number of systematic reviews is growing rapidly, we systematically investigate whether meta-analyses published in leading medical journals present an outline of available evidence by referring to previous meta-analyses and systematic reviews
Characterising and Mitigating Aggregation-Bias in Crowdsourced Toxicity Annotations
No full textTraining machine learning (ML) models for natural language processing usually requires large amount of data, often acquired through crowdsourcing. The way this data is collected and aggregated can have an effect on the outputs of the trained model such as ignoring the labels which differ from the majority. In this paper we investigate how label aggregation can bias the ML results towards certain data samples and propose a methodology to highlight and mitigate this bias. Although our work is applicable to any kind of label aggregation for data subject to multiple interpretations, we focus on the effects of the bias introduced by majority voting on toxicity prediction over sentences. Our preliminary results point out that we can mitigate the majority-bias and get increased prediction accuracy for the minority opinions if we take into account the different labels from annotators when training adapted models, rather than rely on the aggregated labels.Accepted Author ManuscriptWeb Information System
- …
