Network meta-analysis: a technique to gather evidence from direct and indirect comparisons

Area of Science:

• Clinical Medicine
• Epidemiology
• Biostatistics

Background:

• Systematic reviews and pairwise meta-analyses are top-tier evidence but limited to two interventions.
• Many treatments exist, but head-to-head trials are scarce, hindering comprehensive comparisons.
• Traditional methods offer incomplete data for informed clinical and policy decisions.

Purpose of the Study:

• Introduce network meta-analysis (NMA) as an advanced statistical technique.
• Explain NMA's ability to simultaneously compare multiple interventions using direct and indirect evidence.
• Highlight the evolution, assumptions, risks, and benefits of NMA for evidence-based practice.

Main Methods:

• Network meta-analysis (NMA) integrates direct and indirect treatment comparisons within a single model.
• Utilizes various statistical frameworks (Frequentist, Bayesian) and software packages.
• Addresses challenges in NMA conduction, reporting, and interpretation.

Main Results:

• NMA allows simultaneous estimation of effect metrics for all possible treatment comparisons.
• Gathers both direct evidence from head-to-head trials and indirect evidence from mixed treatment comparisons.
• Provides a more complete profile of intervention efficacy and safety.

Conclusions:

• Network meta-analysis offers a powerful tool for evidence-based decision-making when multiple interventions are available.
• Careful consideration of NMA's complexity and assumptions is crucial for reliable results.
• NMA advances the hierarchy of evidence by enabling comprehensive treatment comparisons.