Integrative Analysis of Longitudinal Studies of Aging (IALSA): A Collaborative Network for Parallel and Pooled Data Analysis of Within-Person Change

• Activity Number: 626
• Type: Topic Contributed
• Date/Time: Thursday, August 8, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Andrea Piccinin*+
• Companies: University of Victoria
• Keywords: Reproducibility ; Meta-analysis ; Harmonization ; Coordinated analysis ; Co-calibration ; Longitudinal

Abstract

The IALSA network was formed to facilitate cumulative research on within-person change. Non-experimental longitudinal studies of aging, health and cognition differ markedly in attrition, practice, selection, birth cohort, country, culture, language, and availability of covariates, posing a challenge to comparison of results, measurement harmonization and implementation of meta-analysis. Various consortia have approached these challenges, but lack of measure comparability in currently available data often aggravates these attempts and risks optimistic homogenization of disparate measurements. Coordinated analysis focused on construct level replication has likewise been criticized for potentially leaving the researcher with sets of results that are not easily reconciled. Such analysis, however, reduces variation due to analytic method or inclusion of different covariates seen in the general literature, and is compatible with meta-analysis when measures are comparable. While working on coordinated analyses with construct level reproducibility, we actively pursue opportunities for co-calibration, harmonization, and meta-analysis. Promises, challenges, and synergies will be discussed.