Bayesian Evidence Synthesis in Practice

Date & Time	March 28, 2022, 9:00 a.m. - 4:00 p.m.
Location	University Medical Center Hamburg-Eppendorf
Minimum Number of Participants	10
Maximum Number of Participants	45
Tutorial Fee	100 € / 75 € (regular / student)
Lecturers	Christian Röver, Sebastian Weber

Content

The compilation, summary and processing of information from separate sources is useful in many research areas. The use of Bayesian techniques is a natural option in such meta-analytic endeavours. In medical applications in particular, this allows to derive informative prior distributions, which may then be used to plan future trials with reduced sample sizes, to evaluate their designs, or to support their analysis.

We provide an introduction to the basic methods, the applications in biostatistics, and conduct practical exercises. Meta-analytic methods are introduced starting from the general ideas (like exact and approximate likelihood approaches and hierarchical models) to more advanced issues (like prediction, shrinkage estimation and robustification). Motivated by medical applications, we show how information from historical studies may be used to derive a meta-analytic predictive (MAP) prior, which may then be utilized in a range of ways. In particular, we demonstrate applications making use of historical control data in clinical trials for the purpose of reducing the control group sample size, understanding their design properties, and evaluating the probability of success for a planned future trial. The tutorial includes hands-on exercises using R.

Requirements

Participants should ideally be familiar with the statistical programming language R and bring their laptop, running an R installation as well as the packages "bayesmeta" und "RBesT" (https://cran.r-project.org/package=bayesmeta, https://cran.r-project.org/package=RBesT).

Lecturers

Dr. Christian Röver

Christian Röver is a research associate at the Department of Medical Statistics, University Medical Center Göttingen, Germany. After studying Statistics at Dortmund University and Iowa State University, he earned a PhD degree at The University of Auckland. While his masters thesis was on classification methods, the PhD thesis was on computer intensive methods for Bayesian parameter estimation. After his PhD, he went on to work mostly on signal detection and parameter estimation problems at the Max Planck Institute for Gravitational Physics (Albert Einstein Intitute) in Hannover, before moving to medical statistics at the University Medical Center Göttingen. His current methodological research focus is on Bayesian methods for meta-analysis and their implementation in R.

Dr. Sebastian Weber

Sebastian Weber is working as Director in the Department of Advanced Methodology and Data Science at Novartis. He holds a PhD in Physics from the TU Darmstadt and joined Novartis 8+ years ago. He has worked extensively on enabling the use of historical (control) information in clinical trials through consulting and working on tools to facilitate the application of historical control information from trial design to analysis. Furthermore, Sebastian has experience in designing Oncology phase I dose-escalation trials and is also involved in pediatric drug development programs, where he applies extrapolation concepts. His research interests include the application of pharmacometrics in statistics, model-based drug development and application of Bayesian methods for drug development.

References

C. Röver. Bayesian random-effects meta-analysis using the bayesmeta R package. Journal of Statistical Software, 93(6):1-51, 2020. https://doi.org/10.18637/jss.v093.i06
S. Weber, Y. Li, J. Seaman, T. Kakizume, H. Schmidli. Applying meta-analytic-predictive priors with the R Bayesian evidence synthesis tools. Journal of Statistical Software, 100(19):1-32, 2021. https://doi.org/10.18637/jss.v100.i19