Network synthesis specification
Specify direct and indirect comparisons
The comparison identity explains direction on an additive scale. The analysis still needs model, heterogeneity, multi-arm covariance, uncertainty and diagnostic specifications. Rankings are optional outputs.
Flow diagram
Steps
Define comparison estimands
Specify outcome, time point, effect scale and target population.
Choose model and inference framework
Justify model structure, heterogeneity assumptions and any Bayesian priors.
Specify dependence handling
Model multi-arm correlations and avoid duplicate controls or repeated outcomes.
Plan estimation and diagnostics
Record software, convergence or fit checks, sparse data handling and uncertainty.
Specify interpretable outputs
Plan relative and absolute effects; any ranking includes uncertainty and clinical context.
Network synthesis specification
End flow
Tables and supporting information
Network model plan
Prespecify models; no estimates are calculated here.
Assumptions and rationale
Do not infer a clinically best treatment solely from rank probabilities. Check model fit and direct versus indirect agreement.
Method and protocol evidence
Attach the protocol, methodological sources and versioned software specification supporting these decisions. Record citations and provenance; do not enter participant identifiers.
Indirect comparison notation
d_XY denotes Y relative to X on an additive analysis scale. This consistency identity requires justified transitivity; it is not an automatic calculation.