Workshop on Causal Inference in Health Research

May 9-13, 2011

Program
 

Monday, May 9, 2011
08:30 - 09:30
Registration and Coffee & Croissants
Room(s) 1221, Pavillon André-Aisenstadt


Meeting room(s) : 1140, Pavillon André-Aisenstadt

09:00 - 09:20
Erica E. M. Moodie (McGill University)
Robert Platt (McGill University)
Jay Scott Kaufman (McGill University)
Welcoming addresses

Session - GENERAL TOPICS
Meeting room(s) : 1140, Pavillon André-Aisenstadt

09:25 - 10:00
Andrea Rotnitzky
(Harvard School of Public Health)
Improved double-robust estimation in missing data and causal inference models
Abstract
10:00 - 10:30
Coffee break
Room(s) 1221, Pavillon André-Aisenstadt
10:30 - 11:05
Tyler VanderWeele
(Harvard School of Public Health)
Some results on confounding
Abstract
11:10 - 11:45
Nicholas Jewell
(University of California, Berkeley)
Post-randomization factors, direct effects and randomized trials
Abstract
11:45 - 13:15
Lunch break
13:15 - 14:30
Poster Session
14:30 - 15:00
Coffee break
Room(s) 1221, Pavillon André-Aisenstadt
15:00 - 15:35
Joshua Angrist
(MIT)
ExtrapoLATE-ing: external validity and overidentification in the LATE framework
Abstract
15:40 - 16:15
Elizabeth Stuart
(Johns Hopkins Bloomberg School of Public Health)
From propensity scores to principal scores: matching methods in the context of principal stratification
Abstract

 

Tuesday, May 10, 2011
08:30 - 09:00
Coffee & Croissants
Room(s) 1221, Pavillon André-Aisenstadt

Session - TIME VARYING TREATMENTS AND OPTIMAL TREATMENT STRATEGIES
Meeting room(s) : 1140, Pavillon André-Aisenstadt

09:00 - 09:35
James Robins
(Harvard School of Public Health)
A comparison of causal models
Abstract
09:40 - 10:15
Marshall Joffe
(University of Pennsylvania School of Medicine)
Future ignorability for causal inference for repeated measures outcomes
Abstract
10:15 - 10:45
Coffee break
Room(s) 1221, Pavillon André-Aisenstadt
10:45 - 11:20
Jonathan Schildcrout
(Vanderbilt University)
Analyses of longitudinal, hospital clinical laboratory data with application to blood glucose concentrations
Abstract
11:25 - 12:00
Els Goetghebeur
(Universiteit Gent)
Identifiability issues with structural (nested) mean models for the effect of observed exposures over time
Abstract
12:00 - 13:15
Lunch break
13:15 - 13:50
Miguel Hernan
(Harvard School of Public Health)
When to start cART? Comparing dynamic treatment regimes using the parametric g-formula
Abstract
13:55 - 14:30
Lauren Cain
(Harvard School of Public Health)
When to start cART? Comparing dynamic treatment regimes using inverse probability weighting of dynamic marginal structural models
Abstract
14:30 - 15:00
Coffee break
Room(s) 1221, Pavillon André-Aisenstadt
15:00 - 15:35
Bibhas Chakraborty
(Mailman School of Public Health, Columbia University)
Estimating optimal dynamic treatment regimes with shared decision rules across stages: an extension of Q-learning
Abstract
15:40 - 16:15
Susan Murphy
(University of Michigan)
Inference for dynamic treatment regimes
Abstract

 

Wednesday, May 11, 2011
08:15 - 08:45
Coffee & Croissants
Room(s) 1221, Pavillon André-Aisenstadt

Session - RAMDOMIZED TRIALS AND VACCINE EFFICACY
Meeting room(s) : Z-110, Pavillon Claire McNicoll

08:45 - 09:20
Dylan Small
(The Wharton School, University of Pennsylvania)
Sensitivity analysis for the malaria attributable fever fraction and how collecting symptom information reduces sensitivity to unmeasured confounding
Abstract
09:25 - 10:00
Tom Greene
(The University of Utah, School of Medicine)
Evaluating longitudinal change in randomized trials with attrition due to competing risks
Abstract
10:00 - 10:30
Coffee break
Room(s) 1221, Pavillon André-Aisenstadt
10:30 - 11:05
M. Elizabeth Halloran
(University of Washington)
Causal inference for vaccine effects on infectiousness
Abstract
11:10 - 11:45
Bryan E. Shepherd
(Vanderbilt University)
Sensitivity analysis of per-protocol treatment efficacy in randomized blinded trials
Abstract
11:45 - 13:15
Lunch break
13:15 - 13:50
Michael Hudgens
(University of North Carolina at Chapel Hill)
Randomization-based Inference within principal strata
Abstract
13:55 - 14:30
Dean Follmann
(National Institutes of Health)
Bounds on the effect of vaccine induced immune response on outcome
Abstract

14:30 - 15:00
Coffee break
Room(s) 1221, Pavillon André-Aisenstadt

Session - 2011 AISENSTADT CHAIR “Thematic Semester on Statistics”
Meeting room(s) : Z-110, Pavillon Claire McNicoll

15:00 - 16:00
James Robins
(Harvard School of Public Health)
Modern Mathematical Methods for Drawing Causal Inferences from Observational Data
Abstract

16:00
Wine-and-cheese reception
Room(s) 6245, Pavillon André-Aisenstadt

 

Thursday, May 12, 2011
08:15 - 08:45
Coffee & Croissants
Room(s) 1221, Pavillon André-Aisenstadt

Session - MULTIPLE BIAS MODELING
Meeting room(s) : Z-330, Pavillon Claire McNicoll

08:45 - 09:20
Timothy L. Lash
(Aarhus University Hospital)
Bias analysis to guide new data collection: comprehensive CYP2D6 genotyping in a study of tamoxifen resistance as an example
Abstract
09:25 - 10:00
Eric J. Tchetgen Tchetgen
(Harvard School of Public Health)
Semiparametric theory for Causal Mediation Analysis: robustness, efficiency and sensitivity
Abstract
10:00 - 10:30
Coffee break
Room(s) 1221, Pavillon André-Aisenstadt
10:30 - 11:05
Rich MacLehose
(University of Minnesota School of Public Health)
Is probabilistic bias analysis approximately Bayesian
Abstract
11:10 - 11:45
Paul Gustafson
(University of British Columbia)
Bayesian credible intervals in partially identified models
Abstract

11:45 - 13:15
Lunch break


Meeting room(s) : Z-110, Pavillon Claire McNicoll

13:15 - 13:50
Sara Geneletti
(London School of Economics)
The random discontinuity design in epidemiology
Abstract
13:55 - 14:30
Sebastien Haneuse
(Harvard School of Public Health)
Design-based strategies for bias in observational studies
Abstract
14:30 - 15:00
Coffee break
Room(s) 1221, Pavillon André-Aisenstadt
15:00 - 15:35
Dan Jackson
(University Forvie Site)
Modelling multiple sources of dissemination bias in meta-analysis
Abstract
15:40 - 16:15
Lawrence McCandless
(Simon Fraser University)
Meta-analysis of obsevational data
Abstract

 

Friday, May 13, 2011
08:15 - 08:45
Coffee & Croissants
Room(s) 1221, Pavillon André-Aisenstadt

Session - MISSING DATA METHODS
Meeting room(s) : 1140, Pavillon André-Aisenstadt

08:45 - 09:20
Thomas S. Richardson
(University of Washington)
Null-robust methods for estimating causal effects in randomized trials with non-compliance
Abstract
09:25 - 10:00
Daniel O. Scharfstein
(Johns Hopkins University )
Sensitivity analysis for non-monotone missing data with application to tuberculosis studies
Abstract
10:00 - 10:30
Coffee break
Room(s) 1221, Pavillon André-Aisenstadt
10:30 - 11:05
Jennifer Hill
(New York University)
Flexible accommodation of predictor missing data in causal settings using an extension of a Bayesian nonparametric algorithm
Abstract
11:10 - 11:45
Robin Mitra
(University of Southampton)
Estimating propensity scores with missing covariate data using a latent class mixture model
Abstract
11:45 - 13:15
Lunch break
13:15 - 13:50
Joseph W. Hogan
(Brown University)
Imputation-based inference about natural direct and indirect effects
Abstract
13:55 - 14:30
Mike Daniels
(University of Florida)
Bayesian methods for causal inference on mediators with application to a weight management clinical trial
Abstract
14:30 - 15:00
Coffee break
Room(s) 1221, Pavillon André-Aisenstadt
15:00 - 15:35
Jason Roy
(University of Pennsylvania )
Hierarchical models for causal comparisons of drug effectiveness from administrative data with partially missing confounding variables
Abstract

Last modified : Friday, April 29, 2011 11:21