Session: Methodological Advances in Synthesizing Individual Participant Data Across Multiple Studies (Society for Prevention Research 21st Annual Meeting)

The goal of this symposium is to present new approaches to conducting a research synthesis of individual participant data from multiple studies. Unlike conventional meta-analysis, which relies on summary statistics, all of the presentations illustrate how a researcher may draw substantive conclusions from combining individual participant data from many independent studies into a simultaneous analysis of a single pooled sample.  The symposium supports the conference theme of innovative methods and statistics.

The first paper, presented by Paul Greenbaum (University of South Florida), "Integrated Data Analysis for Longitudinal Data: Pooled Estimates Based on Multiple Calibrations," highlights the advantages of using multiple calibrations for conducting an integrative data analysis of longitudinal studies. The second paper, presented by Wei Wang (University of South Florida), " Multiple Calibrations in Integrative Data Analysis: When Needed, How Many, and How to Combine?" addresses the theoretical and practical aspects of applying multiple calibrations for longitudinal IDA. The third paper, presented by Getachew Dagne (University of South Florida) "Testing Moderation in Network Meta-Analysis with Individual Level Data," examines moderation analysis using individual participant data and shows the power gains that may be obtained by combining trials while taking into account within- and between-trial heterogeneity, and covariates.

At the conclusion of the presentations, the chair will make comments and summary statements and lead a discussion between the presenters and the symposium attendees.