Dataset Information

A Markov chain approach for ranking treatments in network meta-analysis.

ABSTRACT: When interpreting the relative effects from a network meta-analysis (NMA), researchers are usually aware of the potential limitations that may render the results for some comparisons less useful or meaningless. In the presence of sufficient and appropriate data, some of these limitations (eg, risk of bias, small-study effects, publication bias) can be taken into account in the statistical analysis. Very often, though, the necessary data for applying these methods are missing and data limitations cannot be formally integrated into ranking. In addition, there are other important characteristics of the treatment comparisons that cannot be addressed within a statistical model but only through qualitative judgments; for example, the relevance of data to the research question, the plausibility of the assumptions, and so on. Here, we propose a new measure for treatment ranking called the Probability of Selecting a Treatment to Recommend (POST-R). We suggest that the order of treatments should represent the process of considering treatments for selection in clinical practice and we assign to each treatment a probability of being selected. This process can be considered as a Markov chain model that allows the end-users of NMA to select the most appropriate treatments based not only on the NMA results but also to information external to the NMA. In this way, we obtain rankings that can inform decision-making more efficiently as they represent not only the relative effects but also their potential limitations. We illustrate our approach using a NMA comparing treatments for chronic plaque psoriasis and we provide the Stata commands.

SUBMITTER: Chaimani A

PROVIDER: S-EPMC7821202 | biostudies-literature | 2021 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A Markov chain approach for ranking treatments in network meta-analysis.

Chaimani Anna A Porcher Raphaël R Sbidian Émilie É Mavridis Dimitris D

Statistics in medicine 20201026 2

When interpreting the relative effects from a network meta-analysis (NMA), researchers are usually aware of the potential limitations that may render the results for some comparisons less useful or meaningless. In the presence of sufficient and appropriate data, some of these limitations (eg, risk of bias, small-study effects, publication bias) can be taken into account in the statistical analysis. Very often, though, the necessary data for applying these methods are missing and data limitations ...[more]

PMID: 33105517

Similar Datasets

Project description:ObjectiveTo empirically explore the level of agreement of the treatment hierarchies from different ranking metrics in network meta-analysis (NMA) and to investigate how network characteristics influence the agreement.DesignEmpirical evaluation from re-analysis of NMA.Data232 networks of four or more interventions from randomised controlled trials, published between 1999 and 2015.MethodsWe calculated treatment hierarchies from several ranking metrics: relative treatment effects, probability of producing the best value [Formula: see text] and the surface under the cumulative ranking curve (SUCRA). We estimated the level of agreement between the treatment hierarchies using different measures: Kendall's τ and Spearman's ρ correlation; and the Yilmaz [Formula: see text] and Average Overlap, to give more weight to the top of the rankings. Finally, we assessed how the amount of the information present in a network affects the agreement between treatment hierarchies, using the average variance, the relative range of variance and the total sample size over the number of interventions of a network.ResultsOverall, the pairwise agreement was high for all treatment hierarchies obtained by the different ranking metrics. The highest agreement was observed between SUCRA and the relative treatment effect for both correlation and top-weighted measures whose medians were all equal to 1. The agreement between rankings decreased for networks with less precise estimates and the hierarchies obtained from [Formula: see text] appeared to be the most sensitive to large differences in the variance estimates. However, such large differences were rare.ConclusionsDifferent ranking metrics address different treatment hierarchy problems, however they produced similar rankings in the published networks. Researchers reporting NMA results can use the ranking metric they prefer, unless there are imprecise estimates or large imbalances in the variance estimates. In this case treatment hierarchies based on both probabilistic and non-probabilistic ranking metrics should be presented.

Project description:Multiple treatments for early-moderate grade symptomatic haemorrhoids currently exist, each associated with their respective efficacy, complications, and risks. The aim of this study was to compare the relative clinical outcomes and effectiveness of interventional treatments for grade II-III haemorrhoids. A systematic review was conducted according to PRISMA criteria for all the RCTs published between 1980 and 2020; manuscripts were identified using the MEDLINE, Embase, and CENTRAL databases. Inclusion criteria were RCTs comparing procedural interventions for grade II-III haemorrhoids. Primary outcomes of interest were: symptom recurrence at a minimum follow-up of 6 weeks, postprocedural pain measured on a visual analogue scale (VAS) on day 1, and postprocedural complications (bleeding, urinary retention, and bowel incontinence). After bias assessment and heterogeneity analysis, a Bayesian network meta-analysis was performed. Seventy-nine RCTs were identified, including 9232 patients. Fourteen different treatments were analysed in the network meta-analysis. Overall, there were 59 RCTs (73 per cent) judged as being at high risk of bias, and the greatest risk was in the domain measurement of outcome. Variable amounts of heterogeneity were detected in direct treatment comparisons, in particular for symptom recurrence and postprocedural pain. Recurrence of haemorrhoidal symptoms was reported by 54 studies, involving 7026 patients and 14 treatments. Closed haemorrhoidectomy had the lowest recurrence risk, followed by open haemorrhoidectomy, suture ligation with mucopexy, stapled haemorrhoidopexy, and Doppler-guided haemorrhoid artery ligation (DG-HAL) with mucopexy. Pain was reported in 34 studies involving 3812 patients and 11 treatments. Direct current electrotherapy, DG-HAL with mucopexy, and infrared coagulation yielded the lowest pain scores. Postprocedural bleeding was recorded in 46 studies involving 5696 patients and 14 treatments. Open haemorrhoidectomy had the greatest risk of postprocedural bleeding, followed by stapled haemorrhoidopexy and closed haemorrhoidectomy. Urinary retention was reported in 30 studies comparing 10 treatments involving 3116 participants. Open haemorrhoidectomy and stapled haemorrhoidopexy had significantly higher odds of urinary retention than rubber band ligation and DG-HAL with mucopexy. Nine studies reported bowel incontinence comparing five treatments involving 1269 participants. Open haemorrhoidectomy and stapled haemorrhoidopexy had the highest probability of bowel incontinence. Open and closed haemorrhoidectomy, and stapled haemorrhoidopexy were associated with worse pain, and more postprocedural bleeding, urinary retention, and bowel incontinence, but had the lowest rates of symptom recurrence. The risks and benefits of each treatment should be discussed with patients before a decision is made.

Project description:IntroductionAlthough one of the major presentations of vestibular migraine is dizziness with/without unsteady gait, it is still classified as one of the migraine categories. However, in contrast to ordinary migraine, vestibular migraine patients have distinct characteristics, and the detailed treatment strategy for vestibular migraine is different and more challenging than ordinary migraine treatment. Currently, there is no conclusive evidence regarding its management, including vestibular migraine prophylaxis.AimThe objective of this current network meta-analysis (NMA) was to compare the efficacy and acceptability of individual treatment strategies in patients with vestibular migraine.MethodsThe PubMed, Embase, ScienceDirect, ProQuest, Web of Science, ClinicalKey, Cochrane Central, and ClinicalTrials.gov databases were systematically searched for randomized controlled trials (RCTs), with a final literature search date of 30 December 2022. Patients diagnosed with vestibular migraine were included. The PICO of the current study included (1) patients with vestibular migraine; (2) intervention: any active pharmacologic or non-pharmacologic intervention; (3) comparator: placebo-control, active control, or waiting list; and (4) outcome: changes in migraine frequency or severity. This NMA of RCTs of vestibular migraine treatment was conducted using a frequentist model. We arranged inconsistency and similarity tests to re-examine the assumption of NMA, and also conducted a subgroup analysis focusing on RCTs of pharmacological treatment for vestibular migraine management. The primary outcome was changes in the frequency of vestibular migraines, while the secondary outcomes were changes in vestibular migraine severity and acceptability. Acceptability was set as the dropout rate, which was defined as the participant leaving the study before the end of the trial for any reason. Two authors independently evaluated the risk of bias for each domain using the Cochrane risk-of-bias tool.ResultsSeven randomized controlled trials (N = 828, mean age 37.6 years, 78.4% female) and seven active regimens were included. We determined that only valproic acid (standardized mean difference [SMD] -1.61, 95% confidence interval [CI] -2.69, -0.54), propranolol (SMD -1.36, 95% CI -2.55, -0.17), and venlafaxine (SMD -1.25, 95% CI -2.32, -0.18) were significantly associated with better improvement in vestibular migraine frequency than the placebo/control groups. Furthermore, among all the investigated pharmacologic/non-pharmacologic treatments, valproic acid yielded the greatest decrease in vestibular migraine frequency among all the interventions. In addition, most pharmacologic/non-pharmacologic treatments were associated with similar acceptability (i.e. dropout rate) as those of the placebo/control groups.ConclusionsThe current study provides evidence that only valproic acid, propranolol, and venlafaxine might be associated with beneficial efficacy in vestibular migraine treatment.Trial registrationCRD42023388343.

Dataset Information

A Markov chain approach for ranking treatments in network meta-analysis.

Publications

A Markov chain approach for ranking treatments in network meta-analysis.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets