Multivariate meta-analysis using individual participant data.
R D Riley, M J Price, D Jackson, M Wardle, F Gueyffier, J Wang, J A Staessen, I R White
Author Information
R D Riley: Research Institute of Primary Care and Health Sciences, Keele University, Staffordshire, ST5 5BG, UK.
M J Price: School of Health and Population Sciences, Public Health Building, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK.
D Jackson: MRC Biostatistics Unit, Cambridge, UK.
M Wardle: School of Mathematics, Watson Building, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK.
F Gueyffier: UMR5558, CNRS and Lyon 1 Claude Bernard University, Lyon, France.
J Wang: Centre for Epidemiological Studies and Clinical Trials, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Ruijin 2nd Road 197, Shanghai, 200025, China.
J A Staessen: Research Unit Hypertension and Cardiovascular Epidemiology, KU Leuven Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium.
When combining results across related studies, a multivariate meta-analysis allows the joint synthesis of correlated effect estimates from multiple outcomes. Joint synthesis can improve efficiency over separate univariate syntheses, may reduce selective outcome reporting biases, and enables joint inferences across the outcomes. A common issue is that within-study correlations needed to fit the multivariate model are unknown from published reports. However, provision of individual participant data (IPD) allows them to be calculated directly. Here, we illustrate how to use IPD to estimate within-study correlations, using a joint linear regression for multiple continuous outcomes and bootstrapping methods for binary, survival and mixed outcomes. In a meta-analysis of 10 hypertension trials, we then show how these methods enable multivariate meta-analysis to address novel clinical questions about continuous, survival and binary outcomes; treatment-covariate interactions; adjusted risk/prognostic factor effects; longitudinal data; prognostic and multiparameter models; and multiple treatment comparisons. Both frequentist and Bayesian approaches are applied, with example software code provided to derive within-study correlations and to fit the models.
