Causal Estimation of Direct and Indirect Effects in Studies With Clustering or Social Network Features
The focus of this symposium is about methods for estimation of direct and indirect effects in public health research. These methods are broadly applicable to estimate effects in the presence of interference (i.e., one participant’s exposure affects another participant’s outcome). For example, vaccine research and implementation science often involve studies with a natural clustering by social network, community, or school. The direct effect is the effect on the participants who directly received the intervention and the indirect effect is the effect on the participants who shared a cluster with directly exposed participants. This session will cover the state-of-the-art methodology for estimation of direct and indirect effects in randomized trials and observational studies.
Donna Spiegelman, Harvard T.H. Chan School of Public Health
Tyler VanderWeele, Harvard T.H. Chan School of Public Health
Michael Hudgens, University of North Carolina, Chapel Hill
Ashley Buchanan, Harvard T.H. Chan School of Public Health
Sinan Aral, MIT Sloan School of Management