Although aggregate data (AD) from randomised clinical trials (RCTs) are used in the majority of network meta-analyses (NMAs), other study designs (e.g., cohort studies and other non-randomised studies, NRS) can be informative about relative treatment effects. The individual participant data (IPD) of the study, when available, are preferred to AD for adjusting for important participant characteristics and to better handle heterogeneity and inconsistency in the network.
The presenter developed a suite of Bayesian NMA and network meta-regression (NMR) models allowing for cross-design (RCT and NRS) and cross-format (IPD and AD) synthesis, and implemented these models in a new R package crossnma. This package can be used to apply wide range of evidence synthesis methods, including standard meta-analysis and NMA with only RCT that is available in AD formatting.
This Cochrane Learning Live webinar was aimed at anyone interested in performing meta-analysis and network meta-analysis methods using Bayesian models. The session was delivered in June 2023 and below you will find the videos from the webinar together with the accompanying slides to download [PDF].
Part 1: "Crossnma" package and the theory behind it
Part 2: Working example, limitations and future plans
Part 3: Questions and answers
Tasnim Hamza, Postdoctoral fellow, Institute of Social and Preventive Medicine, University of Bern. Tasnim holds a BSc and MSc in Mathematics from Birzeit University, Palestine and a MSc in Biostatistics and Epidemiology from University of Zurich, Switzerland. Recently, she obtained her PhD degree in Biostatistics and Epidemiology from the University of Bern, Switzerland, under the supervision of Prof. Georgia Salanti. Her PhD research focused on extending the arsenal of the current evidence synthesis methods with two main areas of interests in; enhancing the methods of dose-response meta-analysis and dose-response network meta-analysis and developing new methods to enable combining different data formats that come from different study designs.
Guido Schwarzer is a statistician with a Diploma in Statistics from the University of Dortmund, Germany. He is working at the Institute of Medical Biometry and Statistics, Faculty of Medicine and Medical Center - University of Freiburg, Germany. His special interests are statistical computing and statistical methods for pairwise and network meta-analysis. He has co-authored numerous methodological publications and acted as responsible statistician of several systematic reviews published in high-ranking medicine and psychology journals, including 12 Cochrane reviews. He is the maintainer of several R packages for meta-analysis including the widely used R packages meta and netmeta for pairwise or network meta-analysis as well as R package crossnma.