Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 556 - Causal Inference for Complex Data Challenges
Type: Invited
Date/Time: Thursday, August 6, 2020 : 3:00 PM to 4:50 PM
Sponsor: SSC (Statistical Society of Canada)
Abstract #309394
Title: A Model for Effect Modification Using Targeted Learning with Observational Data Arising from Multiple Studies
Author(s): Mireille Schnitzer* and Yan Liu and Dick Menzies and Edward Kennedy and Andrea Benedetti
Companies: Université de Montréal and McGill University and McGill University Health Centre and Carnegie Mellon University and McGill University
Keywords: meta-analysis; causal inference; individual patient data; biostatistics; tmle

When the effect of treatment may vary by individual, precision medicine can be improved by identifying patient covariates to predict the effect at the individual level. One may impose a working model in order to smooth (or summarize) the conditional effect rather than estimate the effect separately for all possible patient subgroups. When working with observational data one must also adjust for all potential confounders of the treatment-outcome relationship, which can be accomplished with propensity score and/or outcome regression modeling. Due to large data requirements, investigators may be interested in using the individual patient data from multiple studies. Our data arise from a systematic review of observational studies contrasting different treatment regimens for patients with multidrug-resistant tuberculosis, where multiple antibiotics are taken concurrently over a long period to cure the infection. We develop a targeted maximum likelihood estimator (TMLE) to fit a marginal structural model representing the treatment effect model in the individual patient data network meta-analytic setting when, for instance, any given treatment may not be observed in all studies.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program