Studies (k): All studies reported, either two-arm or multi-arm
Treatments (n): All treatments that are assessed in at least one of the studies within the network
Pairwise comparisons (m): If m>k, there are multi-arm studies in the network
Designs (d): Specific subset of treatments compared in a study
Estimation of the effect of each treatment compared to the reference treatment. For the estimates, corresponding confidence intervals and test statistics are reported to assess, whether a treatment differs significantly from the reference treatment.Heterogeneity:
Estimated between-study variance, reflects the amount of heterogeneity among the true treatment effects across studies.
Estimated standard deviation of underlying true treatment effects across studies, can be used to describe the distribution of true effects.
Variability may occur between studies (true heterogeneity, tau^2) and within studies (sampling error). I^2 is the percentage of the total variability, that is due to true heterogeneity.
Q_total: Generalized Cochran’s Q statistic to assess heterogeneity
in the whole network
Q_within: Sum of within-design Q statistics for all designs in the network
Q_between: Between-design Q-statistic expressing the heterogeneity between studies with the same design. Heterogeneity between designs indicates design inconsistency (Higgins et al., 2012).
Rücker, G., Krahn, U., Efthimiou, O., & Schwarzer, G. (2021):
netmeta: Network Meta-Analysis using Frequentist Methods. R package
Higgins, J. P., Jackson, D., Barrett, J. K., Lu, G., Ades, A. E., & White, I. R. (2012). Consistency and inconsistency in network meta-analysis: concepts and models for multi-arm studies. Research synthesis methods, 3(2), 98–110. https://doi.org/10.1002/jrsm.1044
The network graph displays the pattern of direct comparisons between the treatments. The treatments are called nodes in graph theory. If there is evidence from direct comparisons between two nodes, these are joined with a line, that is called edge. The width of these edges is proportional to the inverse standard error of the fixed effect model comparing the corresponding two treatments. In other words, thicker lines between two nodes represent more direct evidence of the corresponding comparison. Colored polygons between three and more nodes indicate multi-arm studies.
Rücker, G., Krahn, U., Efthimiou, O., & Schwarzer, G. (2021): netmeta: Network Meta-Analysis using Frequentist Methods. R package version 1.3-0. https://cran.r-project.org/web/packages/netmeta/netmeta.pdf
The forest plot displays the comparisons of all treatments to a common reference treatment. For each comparison, the estimated effect size, as well as the 95 % confidence interval of this estimate is plotted.
Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.2 (updated February 2021). Cochrane, 2021. Available from www.training.cochrane.org/handbook.