Methodological guidance for individual participant data meta-analyses: a systematic review

Area of Science:

• Biostatistics
• Epidemiology
• Health Research Methodology

Background:

• Individual participant data meta-analyses (IPD-MAs) reanalyze original data, offering advantages over traditional meta-analyses but are complex.
• Existing methodological guidance for IPD-MAs is fragmented, particularly for observational studies.
• There is a need for a standardized critical appraisal tool to assess the quality of IPD-MA reports.

Purpose of the Study:

• To systematically identify and synthesize existing methodological guidance for conducting IPD-MAs.
• To inform the development of a critical appraisal tool for IPD-MA reports.
• To identify best practices and critical domains for evaluating IPD-MAs.

Main Methods:

• Systematic literature search of nine electronic databases and gray literature up to June 2025.
• Inclusion of articles reporting empirical, simulation-based, consensus-based, or narrative research with IPD-MA methodology guidance.
• Independent study selection and data extraction by two reviewers, with quality assessment using tailored tools.

Main Results:

• 141 studies were included, encompassing simulation, empirical, and methodological guidance.
• Key themes identified: IPD-MA planning, data access and harmonization, analytical strategies, statistical issues, and reporting.
• Robust guidance exists for randomized trials, but recommendations for observational studies are sparse; 63% of included studies were high quality.

Conclusions:

• The review provides an integrative synthesis of fragmented guidance, highlighting best practices for IPD-MAs.
• Significant gaps in guidance were identified for IPD-MAs of observational studies.
• The synthesized findings serve as the evidence base for developing a dedicated IPD-MA critical appraisal tool.