Bayesian Modeling for Economic Evaluation Using "Real World Evidence"


This chapter presents an example of the use of Bayesian modeling with “real world evidence”. This terminology indicates an increasingly popular body of evidence typically collected in the post-marketing framework and, usually, under observational conditions. Typical examples include population registries, cohort studies, or, more generally “electronic health records”. We focus our attention to the case of statistical modeling for economic evaluation of an intervention on the basis of time-to-event outcomes, where the information provided by the available individual level data (e.g. from a Phase III experimental study) is immature and thus benefits from integration with external (population level) sources of information. We present a case study and discuss the advantages of the Bayesian approach.

In Lesaffre E, Baio G and Boulanger B (eds), Bayesian Methods for Pharmaceutical Research
Gianluca Baio
Gianluca Baio
Professor of Statistics and Health Economics