-----Original Message----- From: Stijn Vansteelandt [mailto:stijn.vansteelandt@ugent.be] Sent: dinsdag 16 september 2014 13:33 To: Olivier Thas Cc: Ann Vanreusel; Beatrijs Moerkerke; Carlos De Wagter; Dirk De Bacquer; Dirk Van den Poel; Els Goetghebeur; Georges Van Maele; Georges De Moor; Gino Verleye; Hubert De Saedeleer; Jean-Pierre Ottoy; Jo Demeester; Luc Duchateau; Marc Van Meirvenne; Maria.Ysebaert Ysebaert; Nele De Belie; René Boel; Ronan Van Rossem; Stefan Van Aelst; Thierry Marchant; Yves Rosseel; Tom Loeys; Lieven Clement; Lieven Clement; Karel Vermeulen; Machteld Varewyck; Bart Jacobs; Johan Steen; Bart Van Rompaye; Tim De Craecker; Kristof De Beuf; Holger Cevallos Valdiviezo; Kristof Vansteelandt; Ludger Goeminne; Sjouke Vandenberghe; Arnout Van Messem; Dominik Sznajder; Joris Meys; Dries Benoit; Gerdie Everaert; Dieter Fiems; Herman Van Oyen; An Vandebosch; Peter Dawyndt; Veronique Storme; Jan De Neve Subject: SEMINAR on 18/9 by Vanessa Didelez on Causal Models for Optimal Dynamic Treatments with Survival Outcomes Dear all, You are cordially invited to the following seminar, which will take place on Thursday September 18 at 12h in Room A3 (De Sterre, S9). Vanessa Didelez, University of Bristol, U.K. Causal Models for Optimal Dynamic Treatments with Survival Outcomes A dynamic treatment is a set of rules that assigns for each time point when a treatment decision needs to be made what this should be, given the patient's history so far. An example in the context of HIV studies is "start treatment when CD4 count first drops below 600". An optimal dynamic treatment is such a set of rules that optimises a criterion reflecting better health for the patient. I will give an overview of the main methods in this context (based on structural models), with particular attention to the case of survival outcomes. For instance we may use the g-computation approach or dynamic marginal structural models - the latter are becoming popular for HIV studies as mentioned above. The particular difficulty with survival outcomes is that marginal and conditional models are typically not compatible. I will discuss and compare approaches, as well as the issue of simulating from structural models; this (hopefully) leads to additional insight and understanding of the models used in this context. Best wishes, Stijn.