Aim of workshop
When randomized trials are not feasible, observational studies carefully analyzed with appropriate causal inference methods are our best bet for informing clinical and societal decisions. As many urgent and important public health questions cannot be ethically or feasibly addressed with randomized trials, it is all the more important that observational studies are appropriately analyzed. Because of limitations with conventional approaches, recent years have seen a huge up-tick in “Mendelian randomization” analyses which leverage genetic variants as natural experiments. While the attraction of Mendelian randomization is unquestionable, there remains confusion in the applied and methodological literature on valid implementation and appropriate interpretation. The purpose of this workshop is to bring together the leading experts in Mendelian randomization specifically and causal inference generally to discuss these themes. Two goals of the workshop include building consensus on best practices in interpretation and implementation of current methods, and developing priorities for future methods development and education.
Program
Tuesday, March 19
9:00-9:15 – Welcome and introductions
9:15-10:45 – Recent developments, upcoming data opportunities – Lead Discussant: George Davey Smith
11:00-12:30 – Interpretation of effect estimates – Lead Discussant: Miguel Hernán
12:30-1:30 – Lunch
1:30-3:00 – Triangulation – Lead Discussant: Deborah Lawlor
3:30-4:30 – Day 1 open discussion
Wednesday, March 20
9:00-10:30 – Robust methods and sensitivity analyses – Lead Discussant: Eric Tchetgen Tchetgen
11:00-12:30 – Failure time models – Lead Discussant: Stijn Vansteelandt
12:30-1:30 – Lunch
1:30-2:30 – Day 2 open discussion
3:00-4:30 – Summary: Setting priorities in research/education – Lead Discussants: Jamie Robins and Sonja Swanson
