Assessing the confidence in network meta-analysis results

We applaud Zhao et al. [1] for attempting to address the comparative efficacy of amiodarone and lidocaine in cardiac arrest using a Bayesian (rather than frequentist) network meta-analytic (NMA) approach. A Bayesian NMA provides a posterior distribution of all relative treatment effects between the interventions in the network and allows for probability statements for a particular outcome such as, “there is X% probability that treatment C is better than B,” or “there is Y% probability that treatment D is the most efficacious out of treatments A,B,C,D, and E.” [2] The Bayesian approach may, therefore, be more appropriate than a Frequentist one when the goal is to make a medical decisio n and predict the outcome under some degree of uncertainty.
Source: The American Journal of Emergency Medicine - Category: Emergency Medicine Authors: Source Type: research