The Use of Multiple Treatment Comparisons in Health Technology Assessment
Indirect and multiple treatment comparison (MTC) approaches to synthesis are logical extensions of more established meta-analysis methods. They have great potential for estimating the comparative effectiveness of multiple treatments using an evidence base of trials which individually do not compare all treatment options. Connected networks of evidence can be synthesized simultaneously to provide estimates of the comparative effectiveness of all included treatments, and a ranking of their effectiveness with associated probability statements. The potential for their use in technology assessment is considerable, allowing a more consistent assessment than simpler alternative approaches. Although such models can be viewed as a logical and coherent extension of standard pair-wise meta-analysis, their increased complexity raises some unique issues with far-reaching implications, concerning how we use data in technology assessment while simultaneously raising searching questions about standard pair-wise meta-analysis. In this presentation we will review pair-wise meta-analysis, indirect and MTC approaches to synthesis, clearly outlining the assumptions they make. Despite increasing numbers of such analyses being published, they have far from universal acceptance and concerns which have been raised regarding the methods will be given an airing and discussed.