Methods for pooled analyses of epidemiologic studies

Abstract

This paper presents a systematic methodology for the pooling and analysis of previously conducted epidemiologic studies. It discusses the methodologic issues to consider and the procedures to follow when conducting a pooled analysis of individual subject level data. I review random and fixed effects models to estimate pooled risks and the methods to examine whether heterogeneity in study-specific effect estimates exists. Analytic strategies presented here can be used to assess sources of heterogeneity, particularly those originating from different study designs and methods. I describe methods for incorporating a qualitative assessment of the study design and data collection methods into the quantitative estimation of the overall effect. Finally, I make recommendations for future pooled analyses of epidemiologic studies.