Summer school: Bayesian methods in health economics
Bayesian methods in health economics
20-24 June 2022, UNIL, Université de Lausanne, Lausanne (Switzerland)
This summer school aims at providing an introduction to Bayesian analysis and Markov Chain Monte Carlo (MCMC) methods using R and MCMC sampling software (such as OpenBUGS and JAGS), as applied to cost-effectiveness analysis and typical models used in health economic evaluations. We will also focus on more recent methods for Probabilistic Sensitivity Analysis including Value of Information calculations. As such, it is intended for health economists, statisticians, and decision modellers interested in the practice of Bayesian modelling and will be based on a mixture of lectures and computer practicals. The emphasis will be on examples of applied analysis: software and code to carry out the analyses will be provided.
Participants are encouraged to bring their own laptops for the practicals. We shall assume a basic knowledge of standard methods in health economics and some familiarity with a range of probability distributions, regression analysis, Markov models and random-effects meta-analysis. However, statistical concepts are reviewed in the context of applied health economic evaluations in the lectures. Lectures and practicals are based on Bayesian Methods in Health Economics (BMHE), The BUGS Book (BB) and Evidence Synthesis for Decision Making in Healthcare (ESDM).
The summer school is hosted in the main campus at the University of Lausanne.
- Introduction to health economic evaluations
- Introduction to Bayesian inference
- Introduction to Markov Chain Monte Carlo in BUGS
- Cost-effectiveness analysis with individual-level data
- Aggregated-level data and hierarchical models
- Evidence synthesis and network meta-analysis
- Model error and structural uncertainty
- Markov models
- Survival analysis
- Missing data in cost-effectiveness modelling
- Introduction to the value of information
- Expected value of partial information (1) - algebraic tricks
- Expected value of partial information (2) - generalised additive models & GP regression
- Expected value of partial information (3) - GP regression in INLA/SPDE
- Expected value of sample information (1) - conjugated analysis
- Expected value of sample information (2) - efficient nested simulation and moment matching
- Expected value of sample information (3) - regression- and sufficient statistics-based methods
Software & useful information
- OpenBUGS (free software for Bayesian analysis)
- R (free general statistical software)
- JAGS (alternative software for Bayesian analysis) - probably the easiest option for Linux or Mac users
- Stan (yet another software for Bayesian analysis) - this is based on a different method for MCMC (called Hamiltonian Monte Carlo)
- R2OpenBUGS (R library to interface R and OpenBUGS) or R2jags (does the same for R and JAGS) or rstan (does the same for R and Stan)
- BCEA (R library to perform Bayesian Cost Effectiveness Analysis). The stable (CRAN) current version is 2.5. The GitHub version is 2.6.
- Wine (a “compatibility layer” that allows to run Windows applications from Linux or Mac)
- Instructions to install OpenBUGS using Wine (for Mac users)
There are two types of registration:
- Students: CHF500 (~£411/ €493)
- Professionals: CHF1,500 (~£1,234 / €1,478)
The registration fee covers
- Tuition fees
- Social events
but please note that accommodation is not covered by the fees.