BUGSnet
BUGSnet is an R package that helps researchers perform and report Bayesian network meta-analyses in a way that meets current scientific and reporting standards. It makes it easier to analyze multiple treatments and present results clearly.
At a glance
Use when
Conducting Bayesian network meta-analyses requiring comprehensive reporting, model diagnostics, and visualization in line with regulatory or guideline standards
Avoid when
Users lack familiarity with R or Bayesian methods; when only frequentist NMA methods are desired; limited computational resources
Inputs
Individual patient data or aggregate-level data (e.g., event counts, means, standard deviations) for multiple interventions within a network
Outputs
Posterior distributions of treatment effects, league tables, SUCRA plots, inconsistency and heterogeneity statistics, convergence diagnostics, and model fit metrics
How it works
BUGSnet is an R package that uses JAGS to conduct Bayesian network meta-analysis (NMA) via a generalized linear model. It provides functions for evidence network description, model estimation, convergence and fit assessment, heterogeneity and inconsistency evaluation, and result visualization (e.g., league tables, SUCRA plots). The package supports comprehensive reporting by aligning outputs with current NMA guidelines.
- HTA domains
- Clinical Effectiveness
- Assumptions
- The network is connected; exchangeability of indirect comparisons holds; the chosen statistical model (e.g., random effects) adequately captures heterogeneity; Bayesian priors are appropriately specified
- Strengths
- Supports full Bayesian inference; integrates with JAGS for flexible modeling; produces outputs aligned with reporting guidelines (e.g., NICE-DSU); includes tools for visualization and model diagnostics
- Limitations
- Requires knowledge of R and Bayesian statistics; depends on JAGS installation; may be computationally intensive for large networks
- Also known as
- Bayesian inference Using Gibbs Sampling to conduct a Network meta-analysis
Questions this answers
- › How do different treatments compare in a network of interventions?
- › What is the relative effectiveness or safety of treatments based on direct and indirect evidence?
- › How consistent is the network meta-analysis model?
- › What is the ranking of treatments based on efficacy or safety?
- › How well does the model fit the data?
- › Are there sources of heterogeneity in the network?
References & sources
Related methods
Similar by meaning
Beta record. Generated from the primary source via AI extraction and independent audit, pending final human review.