The use of fixed study main effects in arm-based network meta-analysis.

Area of Science:

• Biostatistics
• Health Research Methodology

Background:

• Network meta-analysis (NMA) encompasses arm-based and contrast-based methodologies.
• Certain arm-based NMA techniques face criticism for potentially violating the principle of concurrent control by recovering inter-study information.

Purpose of the Study:

• To address criticisms of arm-based network meta-analysis.
• To propose a method for preventing the recovery of inter-study information in arm-based NMA.
• To discuss the advantages of refined arm-based NMA.

Main Methods:

• The study focuses on arm-based network meta-analysis (NMA).
• A key methodological point is the fitting of a fixed main effect for studies.
• This approach is shown to prevent the recovery of inter-study information.

Main Results:

• Fitting a fixed main effect for studies effectively prevents inter-study information recovery in arm-based NMA.
• This modification ensures adherence to the principle of concurrent control.
• The study highlights the benefits of this adjusted arm-based approach.

Conclusions:

• Arm-based network meta-analysis can be conducted while strictly adhering to the principle of concurrent control.
• A fixed main effect for studies is a viable strategy to prevent inter-study information recovery.
• This refinement enhances the validity and reliability of arm-based NMA findings.