Conducting Indirect-Treatment-Comparison and Network-Meta-Analysis Studies
This method helps compare different medical treatments even when they haven't been directly studied together in clinical trials. It uses statistical techniques to link evidence across studies and estimate how one treatment performs relative to others in a network of comparisons.
At a glance
Use when
Comparing multiple interventions with limited direct evidence; informing reimbursement or clinical guidelines; prioritizing treatments based on comparative effectiveness.
Avoid when
When key transitivity assumptions are violated (e.g., differing patient populations or study designs); when the network is highly disconnected or sparse; when only a single direct comparison exists.
Inputs
Evidence from randomized controlled trials organized in a treatment network, including direct pairwise comparisons and patient-level or aggregate outcome data.
Outputs
Relative effect estimates (e.g., odds ratios, mean differences) for all treatment pairs in the network, probability rankings (e.g., SUCRA), and assessment of consistency and heterogeneity.
How it works
A methodological framework developed by the ISPOR Task Force for conducting indirect treatment comparisons (ITC) and network meta-analyses (NMA) to synthesize evidence from multiple randomized controlled trials involving different sets of interventions. It provides guidance on assumptions (e.g., similarity, consistency), statistical models (e.g., Bayesian hierarchical models), network structure, heterogeneity assessment, and reporting standards to support decision-making in health technology assessment.
- HTA domains
- Clinical Effectiveness
- Categories
- Evidence SynthesisIndirect Comparisons
- Assumptions
- Transitivity (comparability across studies), consistency (agreement between direct and indirect evidence), and homogeneity of included studies within treatment comparisons.
- Strengths
- Enables simultaneous comparison of multiple treatments using both direct and indirect evidence; supports decision-making when head-to-head trials are lacking; provides probabilistic treatment rankings and comprehensive evidence synthesis.
- Limitations
- Results depend on quality and comparability of included studies; violations of transitivity or consistency can bias results; complex interpretation and methodological requirements may limit accessibility.
- Also known as
- Indirect Treatment Comparison (ITC), Network Meta-Analysis (NMA), Mixed Treatment Comparison (MTC)
Questions this answers
- › How does a treatment compare to others when no head-to-head trial exists?
- › Which treatment is most effective or safe across a set of alternatives?
- › Are the results consistent across direct and indirect evidence?
- › How should multiple interventions be ranked based on comparative effectiveness?
- › What assumptions must be met for valid indirect comparisons?
- › How should heterogeneity and inconsistency be assessed in a network of studies?
References & sources
Related methods
Similar by meaning
Beta record. Based on the original catalogue summary; primary-source enrichment pending.