logo
tagline
 
Network Meta Analysis

Medical decision making requires the assessment of the relative value of a particular intervention versus all other relevant interventions of choice. The value of an intervention is subject to the available evidence (i.e. quality of study, effect size and associated uncertainty in estimates) for that intervention. Ideally, evidence should be available from a randomized controlled trial (RCT) of sufficient size that compares all interventions of interest simultaneously. However, such a study is often not available because RCTs are often designed for registration purposes and generally only include placebo or one active comparator. Network Meta Analyses (NMA) are a valuable alternative to synthesize evidence when the interest is to compare multiple interventions.

NMA, which is a generalized term covering both indirect treatment comparison (ITC) and mixed treatment comparison (MTC) is an extension of traditional meta-analysis (where all included studies compare the same intervention with the same comparator) by including multiple different pair-wise comparisons across a range of interventions. With NMA, the relative efficacy (or safety) of a particular intervention versus competing interventions can be obtained in the absence of head-to-head evidence; an indirect comparison of two interventions is made via a common comparator. Furthermore, NMA can arguably strengthen the inference by including both direct and indirect effects.

A Bayesian approach to NMA can be considered the method of choice because it allows for a probabilistic interpretation and therefore leads naturally into the decision-making context. The ranking of interventions regarding their ability of providing best outcomes becomes particularly useful.

Mapi Values has performed numerous NMAs which have been used to support reimbursement; many have also been published. Based on this experience we not only have the technical expertise to perform these studies, but are also aware of how best to communicate the findings.