Séminaire du CIMMUL | Philippe Boileau

Découvrir des modificateurs d’effet de traitement dans les données d’essais cliniques / Finding Treatment Effect Modifiers in Clinical Trial Data

Philippe Boileau
Université McGill

Résumé

Heterogeneous treatment effects are driven by treatment effect modifiers, pre-treatment covariates that modify the effect of a treatment on an outcome. These variables at the core of precision medicine are often found as a byproduct of treatment rule estimation. However, using treatment rule estimators to identify treatment effect modifiers in clinical trials where the number of pre-treatment covariates exceeds the number of patients often results in elevated false discovery rates. The many false positives translate to wasted resources when conducting follow-up studies. We resolve these issues by developing a framework for (1) defining treatment effect modifier variable importance parameters applicable to high-dimensional clinical trial data and (2) deriving accompanying causal machine learning estimators with desirable properties.

____

Le séminaire aura lieu au local 3860 du pavillon Alexandre-Vachon et en ligne.

Pour rejoindre la réunion Zoom :
https://ulaval.zoom.us/j/62680136430?pwd=eldBYjdNTG5QR2VxTTFqbVM4UGVRZz09

Meeting ID: 626 8013 6430
Passcode: 693150