Interpreting Results from Bayesian Network Meta-Analyses (Nma): A Guide for Non-Statisticians.

• ORA vs. B>1 indicates a higher response rate in favour of treatment A. • The DIC (Deviance Information Criterion) is commonly used to select fixed vs. random effects model • DIC represents a trade-off between model adequacy and complexity (i.e. complex models, for example a random effect vs a fixed effect model, are penalised) • The model associated with the lowest DIC is selected, considering a 3 to 5-point difference in DIC to be meaningful for model selection. • Fig. 5: Rankogram • Fig. 4: Forest plot – Graphical representation of the ORs and CrIs (graphical illustration of the same data as represented in Table 1)