Project Summary

Meta-analysis synthesizes quantitative information from multiple sources by reconciling differences in order to produce empirical evidence based on a whole body of relevant and sound research. The questions of who needs to be treated and whether some treatments work better for certain participants than others are ideally addressed at the individual participant level (as opposed to the study level) via examining treatment effects by participant characteristic interactions or subgroup analyses. Meta-analysis using individual participant data (IPD) overcomes shortcomings of conventional meta-analysis using study-level summary results. It is recognized as “the gold standard,” particularly for investigating such individual participant-level effects important to patient-centered outcomes research (PCOR) and comparative effectiveness research (CER).

Whereas encompassing all available relevant information is an important consideration of meta-analysis, the existing methods that support individual participant-level meta-analysis are constrained by the assumptions that are untenable for a vast majority of meta-analyses. The team proposes to develop a methodology that permits incorporating all available relevant information without requiring the restrictive assumptions of the existing methods. A unique advantage of the proposed methodology is that it can addresses bias resulting from selective IPD access. Access to IPD is selective if the mechanism granting access to IPD is dependent on the study-level effects under consideration and the studies that provide IPD do not reflect the entire evidence base. The team will compare the proposed methodology with the existing ones for bias and efficiency under both selective and nonselective IPD access. The comparison results will inform the PCORI Methodology Standards, the Cochrane Handbook for Systematic Reviews of Interventions, and other meta-analysis practice guidelines regarding best practices for meta-analysis.

Project Information

Mi-Ok Kim, PhD
The Regents of the University of California, San Francisco
$1,058,322

Key Dates

July 2021
April 2026
2021

Tags

Award Type
State State The state where the project originates, or where the primary institution or organization is located. View Glossary
Last updated: March 16, 2022