Alternatives To Winbugs for Network Meta-Analysis.

xxx • Network meta-analysis (NMA) of clinical trial outcomes is usually based on Bayesian statistics and hence requires software for Monte Carlo Markov chain (MCMC) sampling. • The most common choice of software for NMA is currently WinBUGS, in part because there is a large body of WinBUGS code for NMA in the literature. • However, WinBUGS can be slow and difficult to use – the error messages are often unclear and the language does not allow for vector operations or if/then statements. • This project aimed to identify and evaluate alternatives to WinBUGS for use in Bayesian network meta-analysis. • Specifically we were interested in software that was suitable for applied statistical analysis in the setting of a statistical consultancy.