Multi-parameter Evidence Synthesis Research

This programme of work has been underway in since 2002, initially funded by the Medical Research Council’s Health Services Research Collaboration, and then transferred to University of Bristol in 2009. It continues with grant and fellowship funding from Medical Research Council and National Institute of Health Research .

Multi-Parameter Evidence Synthesis (MPES) is a generalization of meta-analysis, broadened and extended so that it can combine not only evidence from different sources and different research designs, but also evidence that informs different functions of parameters.

The approach builds on the Confidence Profile Method of David Eddy and his colleagues in the early 1990’s, but we have extended this to richer, more complex, and more realistic applications. Also, unlike the CPM work, we have emphasized the importance of consistency in the information coming from different evidence sources. The use of the Bayesian Markov chain Monte Carlo software, WinBUGS, has greatly facilitated our work

The programme of research is strongly oriented to a medical decision making context and to economic evaluation, but some of the syntheses we have carried out are exercises in descriptive epidemiology.