A node is as good as its adjoining edge: Bayesian approach to a network of data
*Meg Gamalo, FDA 


In clinical trials, the treatment response to a particular intervention only becomes meaningful when it is compared to another intervention, even if that intervention comes from historical trials. If there is a set of interventions, the number of comparisons forms a network. In this talk, we will discuss a special (undirected) form of Bayesian Networks (BN) called Bayesian network meta-analytic techniques that allow for an assessment of the relative safety/effectiveness of two treatments when they have not been compared directly in a randomized trial, but have each been compared indirectly with other treatments. These methods will be illustrated using real examples found in the literature.