Not a giant leap, but a small step from pair-wise to network meta-analysis
Outline of workshop:
Target audience: Mixed – Cochrane review authors, contributors, managing editors and researchers, who are interested in learning the key concepts involved in the design and undertaking of a Cochrane review with multiple treatment comparisons.
Level of difficulty (basic/intermediate/advanced): Intermediate
Type of workshop (training/discussion): Training
Transitioning from pairwise meta-analysis to network meta-analysis (NMA) is a necessary step for Cochrane, as seen by the addition of new content related to NMA in the 2019 version of the Cochrane handbook for systematic reviews of interventions. This transitioning is not a giant leap, but a small step: NMA allows the comparison of more than two interventions, however, it does not differ from pairwise meta-analysis in its core nature. The primary barriers for implementing NMA in the Cochrane community is the lack of understanding of the principles and methods, and the unavailability of the statistical routines within RevMan. MetaInsight is a freely available interactive online application requiring no specialist statistical software. It conducts NMA, including providing network plot, forest plot, pooled estimates from both pairwise and network evidence, inconsistency tests and sensitivity analysis. MetaInsight can be accessed at: https://crsu.shinyapps.io/metainsightc/
At the end of the course, through problem-based learning, participants should be able to:
• Gain a good understanding of principles of NMA and assess the underlying assumptions of NMA
• Conduct appropriate NMA using MetaInsight
This workshop will explore the principles and the practice of conducting meta-analysis of multiple treatment comparisons. Course participants will gain hands-on experience of conducting pairwise and NMA using MetaInsight. The course is holistic in covering the whole process from the strategic analysis of the decision problem, through conduct of meta-analysis, to synthesis and interpretation of results. During this workshop, we will:
i. Provide participants a good understanding of the basic principles and underlying assumptions of pairwise and network meta-analysis;
ii. Demonstrate the wide array of functions available within MetaInsight, including loading data, generating network plot, visualising data, conducting random effect / fixed effect NMA, generating forest plot from pooled estimates, generating comparison tables containing both pairwise and network estimates for all possible comparisons, conducting and interpreting tests of inconsistency, and exporting tables and figures;
iii. Provide participants with an example dataset and the opportunity to work at their own pace, through a worksheet designed to guide them through the above functions and stages of conducting an analysis in MetaInsight.
Time will be given for questions and discussions. All participants must bring a laptop (or may share with another participant).
Date and time:
Tuesday 21st April, 12:00pm-13.30pm.
Olivia Wu, Neil Hawkins, Suzanne Freeman & Yiqiao Xin
Olivia Wu is the Director of the Health Economics and Health Technology Assessment (HEHTA) Research Unit, at the University of Glasgow and Director of the National Institute for Health Research (NIHR) Complex Reviews Support Unit (CRSU). Olivia's research interest focuses on the areas of evidence synthesis (including systematic review and meta-analysis of aggregate and individual patient data, and of direct, indirect and networked evidence), and the use of evidence synthesis in the context of economic evaluation and health technology assessment (HTA). In addition to her research, Olivia serves as a member of NIHR Systematic Review Programme Advisory Group (since 2015), the NIHR HTA General Board (since 2016) and the Appointment Panel for NIHR Senior Investigator (2018).
Neil Hawkins is Professor of Health Economics & Health Technology Assessment, HEHTA, University of Glasgow and Deputy Director of the National Institute for Health Research (NIHR) Complex Reviews Support Unit (CRSU). Neil holds a PhD in Pharmacology from the University of Bristol, Masters Degrees in Health Economics (York) and Applied Statistics (Sheffield), and also an MBA from the University of Oxford. He has over 17 years’ experience in the field of HTA and has participated in over thirty health technology assessments covering a wide variety of clinical areas. Previously he was a Reader in Health Technology Assessment (HTA) at the London School of Hygiene and Tropical Medicine, and Vice President leading the global Health Economics practice at ICON plc. He has published numerous articles in international journals including Health Economics, Medical Decision Making, Heart and Value in Health. In addition he has conducted workshops discussing methods for evidence synthesis and decision analytic modelling in North and South America, Europe, Asia and Saudi Arabia, and also taught on short courses entitled ‘Introduction to Statistics’, ‘Introduction to Systematic Review and Meta-Analysis’ and ‘Development of Conceptual Models’ at meetings of the International Society for Pharmacoeconomics and Outcomes Research (ISPOR)
Suzanne Freeman is an NIHR Research Fellow at the University of Leicester. Suzanne completed her PhD on individual participant data network meta-analysis of time-to-event outcomes at University College London in 2016. Suzanne has been a member of CRSU since November 2016 and her research interests include network meta-analysis, individual participant data meta-analysis, synthesis of time-to-event and continuous outcomes and diagnostic test accuracy meta-analysis.
Yiqiao Xin is a research associate in HEHTA, University of Glasgow. She joined the group in September 2013. Yiqiao holds a Bachelor degree in Pharmacy (2010) from China Pharmaceutical University, a Masters degree in Health Technology Assessment and Management (2013) from University of Montreal and a PhD in Health Economics (2018) from University of Glasgow. Yiqiao's methodological interests are economic evaluations alongside clinical trials, decision analytic modelling, and measuring and valuing health. She has undertaken economic evaluations within a wide variety of health care and public health areas including Parkinson’s, diabetes, physical activity interventions, and child’s social and emotional wellbeing interventions. Meantime, she is a member of the NIHR Complex Review Support Unit (CRSU) and provides methodology advice on complex methodologies used in NIHR funded projects, as well as co-developing the 'MetaInsight' freely available web-based tool for conducting network meta-analysis.