Title: Improved Monte Carlo inference methods for network meta-analysis
Authors: Hisashi Noma - The Institute of Statistical Mathematics (Japan) [presenting]
Kengo Nagashima - The Institute of Statistical Mathematics (Japan)
Abstract: Network meta-analysis enables comprehensive synthesis of evidence concerning multiple treatments and their simultaneous comparisons based on both direct and indirect evidence. In the practice of network meta-analyses, multivariate random effects models have been routinely used for addressing between-studies heterogeneities. Although their standard inference methods depend on large sample approximations (e.g., restricted maximum likelihood [REML] estimation) for the number of trials synthesized, the numbers of trials are often moderate. In these situations, standard estimators cannot be expected to behave in accordance with asymptotic theory; in particular, confidence intervals cannot be assumed to exhibit their nominal coverage probabilities (also, the type-I error probabilities of the corresponding tests cannot be retained). The invalidity issue may seriously influence the overall conclusions of network meta-analyses. We provide permutation-based Monte Carlo inference methods that enable exact joint inferences for average outcome measures without large sample approximations. We also provide accurate marginal inference methods under general settings of network meta-analyses. We propose effective approaches for permutation inferences using optimal weighting based on the efficient score statistic. The effectiveness of the proposed methods is illustrated via applications to a network meta-analysis for antihypertensive drugs on incident diabetes.