How does the skill help with inconsistency assessment?

It provides code for global inconsistency tests (design-by-treatment interaction) and local inconsistency assessment through node-splitting methods.

What R packages does this skill support for NMA?

The skill supports leading R packages including netmeta for frequentist analysis, gemtc for Bayesian analysis via JAGS, and multinma for Bayesian analysis via Stan.

Does it support treatment ranking visualization?

Yes, it includes implementations for P-scores, SUCRA plots, and rankograms to visualize treatment hierarchy and uncertainty.

Can I perform both contrast-based and arm-based meta-analysis?

Yes, the skill provides patterns for handling both pairwise (contrast-based) data and arm-level data, including conversion functions.

R Network Meta-Analysis Toolkit

Name: R Network Meta-Analysis Toolkit
Author: choxos

bychoxos

Data Science & ML

Performs comprehensive network meta-analyses in R using frequentist and Bayesian frameworks to compare multiple treatments simultaneously.

About

This skill provides specialized patterns and code implementations for performing Network Meta-Analysis (NMA) within the R ecosystem. It bridges the gap between direct and indirect evidence, allowing users to assess network structures, evaluate consistency, and generate treatment rankings. By integrating popular packages like netmeta, gemtc, and multinma, the skill assists researchers in executing complex statistical models, visualizing network geometry, and conducting robust heterogeneity assessments for comparative effectiveness research.

Key Features

Bayesian NMA workflows for both gemtc (JAGS) and multinma (Stan)
0 GitHub stars
Comprehensive network visualization and structure assessment tools
Consistency evaluation via node-splitting and net heat plots
Frequentist NMA implementation using the netmeta package
Advanced treatment ranking techniques including SUCRA and rankograms

Use Cases

Synthesizing evidence from multiple clinical trials to compare three or more medical interventions
Assessing the consistency between direct and indirect treatment comparisons in a network
Generating probabilistic rankings of treatments for clinical decision-making

About

Key Features

Bayesian NMA workflows for both gemtc (JAGS) and multinma (Stan)
0 GitHub stars
Comprehensive network visualization and structure assessment tools
Consistency evaluation via node-splitting and net heat plots
Frequentist NMA implementation using the netmeta package
Advanced treatment ranking techniques including SUCRA and rankograms

Use Cases

Synthesizing evidence from multiple clinical trials to compare three or more medical interventions
Assessing the consistency between direct and indirect treatment comparisons in a network
Generating probabilistic rankings of treatments for clinical decision-making