Table 5 Table of characteristics of methods studies related to NMA biases (n = 27)
First author, year of publication | Title | Objective | Research Institute | Country of the first author | Specific topic area | Methods used |
|---|---|---|---|---|---|---|
Cameron 2017 [67] | Importance of Considering Differences in Study Design in Network Meta-analysis: An Application Using Anti-Tumor Necrosis Factor Drugs for Ulcerative Colitis | Present approaches to adjust findings derived from adaptive study designs to make them more comparable with findings derived from conventional parallel design RCTs | Cornerstone Research Group | Canada | Adaptive trial design RCT | Case study |
Cameron 2018 [68] | Importance of assessing and adjusting for cross-study heterogeneity in network meta-analysis: a case study of psoriasis | Present adjustment for cross-study heterogeneity when conducting NMAs using a case study of biologic therapies for moderate-to-severe plaque psoriasis | Cornerstone Research Group | Canada | No | Case study |
Davies 2021 [69] | Degree irregularity and rank probability bias in network meta-analysis | Study how the structure of the network affects the probability that each treatment is ranked first, second and so on | University of Manchester | UK | No | Case study |
Donegan 2013 [42] | Assessing key assumptions of network meta-analysis: a review of methods | Review and illustrate methods to assess homogeneity and consistency | University of Liverpool | UK | No | Review and case study |
Efthimiou 2020 [70] | The dark side of the force: Multiplicity issues in network meta-analysis and how to address them | Present problems with the usual standard models for NMA in estimating multiple relative treatment effects, and suggest alternative modelling methods | University of Bern | Switzerland | No | Review of the literature |
Efthimiou 2016 [43] | GetReal in network meta-analysis: a review of the methodology | Present methods for NMA and discuss conceptual and statistical ways of evaluating the underlying assumptions of the model while providing guidance for researchers | University of Ioannina School of Medicine | Greece | No | Theoretical arguments and simulations |
Efthimiou 2017 [71] | Multivariate extension of meta-analysis | Objective 1. Provide a comprehensive account of the currently available methods. Objective 2. Advance the statistical methodology for jointly analyzing multiple correlated outcomes in NMAs | University of Ioannina | Greece | No | Review and case study |
Goring 2016 [72] | Disconnected by design: analytic approach in treatment networks having no common comparator | Describe disconnected networks, present alternative Bayesian models and provide a framework to guide the choice between modeling approaches | ICON plc | Ireland | Disconnected networks | Theoretical arguments and simulations |
Jackson 2017 [73] | Paule-Mandel estimators for network meta-analysis with random inconsistency effects | Discuss and simulate Paule-Mandel estimators for NMA with random inconsistency effects | MRC Biostatistics Unit | UK | Random inconsistency effects | Theoretical arguments and simulations |
Jansen 2011 [12] | Is network meta-analysis as valid as standard pairwise meta-analysis? It all depends on the distribution of effect modifiers | Compare pairwise meta-analysis with network meta-analysis with a specific focus on the primary role of effect modifiers as the common cause of heterogeneity and bias | Mapi Group | USA | No | Theoretical arguments |
Kibret 2014 [74] | Bias in the identification of the best treatment in a Bayesian network meta-analysis for binary outcome: a simulation study | Investigate how rank probabilities obtained from a Bayesian NMA are affected by characteristics of the network, including network configuration, number of studies per comparison, individual study sample sizes, and effect sizes | McMaster University | Canada | Bayesian framework | Theoretical arguments and simulation |
Krahn 2013 [75] | A graphical tool for locating inconsistency in network meta-analyses | Provide a tool, the net heat plot, to render transparent which direct comparisons drive each network estimate and to display hot spots of inconsistency | Johannes Gutenberg University | Germany | No | Theoretical arguments and case studies |
Lin 2016 [76] | Sensitivity to Excluding Treatments in Network Meta-analysis | Examine the sensitivity to treatment exclusion of an alternative approach to network meta-analysis, namely the arm-based approach | University of Minnesota | USA | Arm-based approach | Theoretical arguments and case study |
Linde 2016 [77] | Questionable assumptions hampered interpretation of a network meta-analysis of primary care depression treatments | Evaluate the underlying assumptions of a network meta-analysis and highlight challenges and pitfalls of interpretation under consideration of the assumptions | Technische Universitat Munchen | Germany | Effectiveness and acceptability of pharmacologic and psychological treatments for primary care patients with depression | Case study |
Marks-Anglin 2020 [78] | A historical review of publication bias | Offer an historical account of seminal contributions, with an emphasis on the parallel development of graph-based and selection model approaches | University of Pennsylvania | USA | No | Review and theoretical arguments |
Naci 2014 [79] | Industry sponsorship bias in research findings: a network meta-analysis of LDL cholesterol reduction in randomised trials of statins | Explore the risk of industry sponsorship bias | London School of Economics and Political Science | UK | Placebo-controlled and active comparator trials of statins | Case study |
Owen 2020 [80] | Multivariate network meta-analysis incorporating class effects | Extend the trivariate NMA model to incorporate the exchangeability between class treatment effects, and a missing data framework to estimate uncertainty in trials that did not report measures of variability | University of Leicester | UK | No | Theoretical arguments and case studies |
Papakonstantinou 2020 [81] | In network meta-analysis most of the information comes from indirect evidence: empirical study | Examine the relative contribution of network paths of different lengths to estimates of treatment effects | University of Bern | Switzerland | No | Meta-research study of NMAs |
Phillippo 2019 [24] | Threshold Analysis as an Alternative to GRADE for Assessing Confidence in Guideline Recommendations Based on Network Meta-Analyses | Outline threshold analysis as an alternative approach, demonstrating the method with two examples of clinical guidelines | NICE Guidelines Technical Support Unit | UK | Social anxiety and depression guidelines | Theoretical arguments and case studies |
Salanti 2009 [82] | A case study of multiple-treatments meta-analysis demonstrates that covariates should be considered | Illustrate the simultaneous analysis of a network of trials, using as a case study | MRC Biostatistics Unit | UK | Topical fluoride treatments | Case study |
Shi 2018 [83] | Node-making processes in network meta-analysis of nonpharmacological interventions should be well planned and reported | Describe four ways to create a network of nodes based on NMA objectives | University of Manchester | UK | Nonpharmacological interventions | Theoretical argument |
Song 2003 [84] | Validity of indirect comparison for estimating efficacy of competing interventions: empirical evidence from published meta-analyses | Determine the validity of adjusted indirect comparisons by using data from published meta-analyses of randomised trials | University of Birmingham | UK | No | Meta-research study of NMAs |
Stevens 2018 [44] | A review of methods for comparing treatments evaluated in studies that form disconnected networks of evidence | Review and discuss methods for comparing treatments evaluated in studies that form disconnected networks of evidence | University of Sheffield | UK | Disconnected networks | Review of methods and applications |
Swallow 2020 [85] | Causal inference and adjustment for reference-arm risk in indirect treatment comparison meta-analysis | Outline methods to reduce biases associated with ITC/NMA and apply them to three real-world examples | Analysis Group Inc | USA | No | Theoretical arguments and case studies |
Tan 2013 [86] | Presentational approaches used in the UK for reporting evidence synthesis using indirect and mixed treatment comparisons | Provide recommendations to improve indirect comparison/mixed treatment comparison reporting and identify research priorities for improved presentation | University of Leicester | UK | No | Review of guidance |
Thorlund 2013 [87] | Why the findings of published multiple treatment comparison meta-analyses of biologic treatments for rheumatoid arthritis are different: an overview of recurrent methodological shortcomings | Evaluate the quality of published MTCs and to identify methodological issues that can explain the discrepancies in the findings of these MTCs | McMaster University | Canada | Biologic disease-modifying antirheumatic drugs for rheumatoid arthritis | Meta-research study of NMAs |
Tonin 2019 [88] | Description of network meta-analysis geometry: A metrics design study | We aimed to propose metrics adapted from graph theory and social network-analysis literature to numerically describe NMA geometry | Federal University of Parana | Portugal | Any drug therapy intervention | Meta-research study of NMAs |