← ClaudeAtlas

nma-methodologylisted

Deep methodology knowledge for network meta-analysis including transitivity, consistency assessment, treatment rankings, and model selection. Use when conducting or reviewing NMA.
choxos/BiostatAgent · ★ 4 · AI & Automation · score 75
Install: claude install-skill choxos/BiostatAgent
# Network Meta-Analysis Methodology Comprehensive methodological guidance for conducting rigorous network meta-analysis following NICE DSU and PRISMA-NMA guidelines. ## When to Use This Skill - Planning a network meta-analysis - Assessing transitivity and consistency - Interpreting treatment rankings - Choosing between frequentist and Bayesian NMA - Designing NMA sensitivity analyses - Reviewing NMA code or results ## Fundamental Assumptions ### 1. Transitivity Assumption **Definition**: If we can estimate A vs B directly and B vs C directly, we can estimate A vs C indirectly, provided the studies are sufficiently similar. **Requirements**: - Studies comparing different treatments should be similar enough to have been included in the same RCT - Effect modifiers should be balanced across comparisons - No important differences in study-level characteristics **Assessment**: ``` For each comparison in network, check: ├── Population similarity │ - Age, sex, disease severity │ - Biomarker status, prior treatments ├── Outcome definitions │ - Same definition of response/event │ - Same time point of assessment ├── Treatment definitions │ - Dose, duration, route │ - Concomitant medications └── Study design - Randomization, blinding - Follow-up duration ``` **Presenting Transitivity Assessment**: - Create table of study characteristics by comparison - Highlight any systematic differences - Use forest plots stratified by comparison ### 2. Consistency Assumpt