An Alternative Pseudolikelihood Method for Multivariate Random-Effects Meta-Analysis When the Within-Study Correlations Are Unknown

• Activity Number: 692
• Type: Contributed
• Date/Time: Thursday, August 8, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Chuan Hong*+ and Yong Chen and Richard Riley
• Companies: The University of Texas School of Public Health and The University of Texas School of Public Health and University of Birmingham
• Keywords: Composite likelihood ; Correlation ; Multivariate meta-analysis ; Nonconvergence problem ; Pseudolikelihood

Abstract

Recently, multivariate random-effects model in meta-analysis has received a great deal of attention because of its simplicity and ability to account for the within-study and between-study correlations. However, the standard inference procedures, such as the maximum likelihood or maximum restricted likelihood inference, require within-study correlations, which are usually unavailable. In addition, the standard inference procedures suffer from the nonconvergence problem due to the estimate of between-study correlation being close to the boundary of its parameter space. In this paper, we propose a pseudolikelihood method to overcome the aforementioned problems. The pseudolikelihood method does not require within-study correlations, and is not prone to the nonconvergence problem. In addition, it can properly estimate the covariance between pooled estimates for different outcomes, which enables valid inference on functions of pooled estimates, and can be applied to the studies where some of the multiple outcomes are missing completely at random. Simulation studies show that the pseudolikelihood method performs well in finite sample. We illustrate our method through three meta-analyses.