Recent work has made important contributions in the development of causally-interpretable meta-analysis. These methods transport treatment effects estimated in a collection of randomized trials to a target population of interest. Ideally, estimates targeted toward a specific population are more interpretable and relevant to policy-makers and clinicians. However, between-study heterogeneity not arising from differences in the distribution of treatment effect modifiers can raise difficulties in synthesizing estimates across trials. The existence of such heterogeneity, including variations in treatment modality, also complicates the interpretation of transported estimates as a generic effect in the target population. We propose a conceptual framework and estimation procedures that attempt to account for such heterogeneity, and develop inferential techniques that aim to capture the accompanying excess variability in causal estimates. This framework also seeks to clarify the kind of treatment effects that are amenable to the techniques of generalizability and transportability.
Causally-interpretable meta-analysis using aggregate data
University of Groningen（格罗宁根大学）；University Medical Center Utrecht（乌得勒支大学医学中心）；University Medical Center Groningen（格罗宁根大学医学中心）；Groningen Research Institute of Pharmacy（格罗宁根药学研究所）；Dutch National Institute of Public Health and the Environment (RIVM)（荷兰国家公共卫生与环境研究所（RIVM））；Harvard T.H. Chan School of Public Health（哈佛大学陈曾熙公共卫生学院）
Causally Interpretable Meta-Mediation Analysis With Missing At Random Mediator and Outcome Data
EPILOGY, Institut Mondor of Biomedical Research, INSERM U955（INSERM U955 蒙多尔生物医学研究所 EPILOGY）；Université Paris-Est Créteil（巴黎东克雷泰伊大学）；Mines Paris, Université Paris Sciences Lettres（巴黎矿业学院，巴黎文理研究大学）；Department of Mathematics, Université du Québec à Montréal（蒙特利尔大学数学系）
Causal mediation in cluster-randomized trials with multiple mediators: spillover-aware decomposition, identification, and semiparametric efficient inference
