ml-nmr-methodology

ML-NMR Methodology

Comprehensive methodological guidance for conducting rigorous Multilevel Network Meta-Regression following NICE DSU guidance and multinma package documentation.

When to Use This Skill

• Deciding whether ML-NMR is appropriate
• Setting up integration points for AgD
• Specifying priors and models
• Understanding marginal vs conditional effects
• Predicting to target populations
• Reviewing ML-NMR code or results

When to Use ML-NMR

ML-NMR is Appropriate When:

1. Network Structure

   • Multiple treatments form (partial) network
   • Some studies have IPD, others only AgD
   • Want to leverage all available evidence

2. Population Differences

   • Effect modifiers differ across populations
   • Standard NMA transitivity violated
   • Need population-adjusted estimates

3. Target Population

   • Want predictions for specific population
   • Different from any single trial population
   • Policy-relevant population definition

ML-NMR vs Alternatives

Scenario	Recommended Method
All AgD, similar populations	Standard NMA
All AgD, different populations	NMA meta-regression
IPD for one study, AgD for one	MAIC or STC
IPD + AgD network	ML-NMR
Disconnected with IPD	ML-NMR (with assumptions)

Key Concepts

Individual-Level vs Study-Level

ML-NMR Models Both:
├── Individual-level (within IPD studies)
│   - Patient-level outcomes
│   - Patient-level covariates
│   - Exact covariate-outcome relationships
│
└── Study-level (for AgD studies)
    - Aggregate outcomes
    - Covariate summaries
    - Integration over covariate distribution

Population Adjustment

Problem: AgD studies provide aggregate summaries, but we need individual-level predictions.

Solution: Numerical integration over the AgD population's covariate distribution.

For AgD study:
Expected outcome = ∫ f(outcome | covariates, treatment) × p(covariates) d(covariates)

Where:
- f(): Individual-level outcome model (from IPD)
- p(): Covariate distribution in AgD population