itc-analysis

ITC Analysis - Multi-Agent Evidence Synthesis Orchestration

Design and execute a complete indirect treatment comparison analysis for: $ARGUMENTS

Thinking

This workflow orchestrates multiple specialized agents to conduct rigorous ITC analysis following NICE DSU guidance and tidy modelling principles. The approach emphasizes:

• Phase-based coordination: Each phase builds upon previous outputs with clear handoffs
• Method-appropriate specialists: Matching the right expert to the analytical task
• Assumption transparency: Explicit checking and documentation of all assumptions
• Reproducibility: Complete documentation for replication
• Sensitivity analysis: Multiple approaches to validate findings

The multi-agent approach ensures each aspect is handled by domain experts:

• ITC Architect selects methodology
• Appropriate specialists execute analysis
• Code Reviewer validates implementation

Phase 1: Evidence Assessment & Method Selection

Deliverables:

1. Evidence mapping:

   • Available studies and their characteristics
   • IPD vs AgD availability
   • Network structure and connectivity
   • Treatment comparisons of interest

2. Population assessment:

   • Key population characteristics across studies
   • Potential effect modifiers
   • Population differences requiring adjustment

3. Assumption evaluation:

   • Transitivity assessment (for NMA)
   • Effect modifier identification (for MAIC/STC)
   • Data requirements for each method

4. Method recommendation:

   • Primary recommended method with justification
   • Alternative methods for sensitivity
   • Package recommendations (meta, netmeta, maicplus, stc, multinma)

5. Analysis plan outline:

   • Key analyses to perform
   • Sensitivity analyses needed
   • Reporting requirements

Provide clear rationale for method selection referencing NICE DSU TSD guidance.