Does it matter to have study-level or participant-level data when assessing the effect of moderation in a network meta-analysis?
Getachew Dagne, Ph.D.
University of South Florida
Meta-analytic methods for combining data from multiple intervention trials are commonly used to estimate the eﬀectiveness of an intervention. They can also be extended to study comparative eﬀectiveness, testing which of several alternative interventions isexpected to have the strongest eﬀect. This often requires network meta-analysis (NMA), which combines trials involving direct comparison of two interventions within the same trial and indirect comparisons across trials. In this paper, we extend existing network methods for main eﬀects to examining moderator eﬀects, allowing for tests of whether intervention eﬀects vary for diﬀerent populations or when employed in diﬀerent contexts. In addition, we study how the use of individual participant data (IPD) may increase the sensitivity of NMA for detecting moderator eﬀects, as compared to traditional NMA that employs study-level eﬀect sizes in a meta-regression framework. A new network meta-analysis diagram is proposed. We also develop a generalized multilevel model for NMA that takes into account within- and between-trial heterogeneity, and can include individual level covariates. Within this framework we present deﬁnitions of homogeneity and consistency across trials. A simulation study based on this model is used to assess eﬀects on power to detect both main and moderator eﬀects. Results show that power to detect moderation is substantially greater when applied to IPD as compared to study-level eﬀects. We illustrate the use of this method by applying it to data from a classroom-based randomized study that involved two subtrials, each comparing interventions that were contrasted with separate control groups.