Causal Inference & Causal Diagrams in MDM Using Big Real World Observational Data & Pragmatic Trials
Course Description and Objectives:
This course will provide an introduction to the principles of causation and causal diagrams, with focus on Directed Acyclic Graphs (DAG) and a brief introduction to methods for causal inference (“g-methods) including multivariate analysis, propensity scores, g-formula, marginal structural models with inverse probability of treatment weighting, and structural nested models with g-estimation (lecture - exercises - discussion). We will use the “target trial” concept and a counterfactual approach with “replicates” to apply causal methods and valid per-protocol analysis of sustained treatment regimens to big real world observational data and pragmatic trials with postrandomization confounding.
The objectives of this course are to draw and interpret causal diagrams, to decide which biostatistical/epidemiological methods must be used in different situations to derive causal effect parameters, to understand causal modeling, and to know how these methods are applied in large real world data and pragmatic trials.
Published cardiovascular, oncology, HIV, nutrition and obstetrics examples will be used to: *** Adjust for non-adherence in randomized clinical trials, where both the intention-to-treat and the naïve per protocol analyses can fail to yield the true causal intervention effect; *** Assess the “fallibility of estimating direct effects” (i.e., adjusting for intermediate steps); *** Adjust for time-independent confounders (i.e., confounder affects both risk factor and disease), where standard stratification, regression analysis or propensity score methods yield valid causal effects if all confounders are measured, and *** Adjust for time-dependent confounding (i.e., the confounder simultaneously acts as an intermediate step in the causal chain between risk factor and disease), where standard regression analysis fails and causal methods such as marginal structural models or g-estimation must be used, and finally, *** Emulate randomized controlled trials and standardize/transport estimates to the trial population for validation purposes.
Register | Return to Short Course Listings
Become a Member
SMDM members contribute critically to health policy research in the areas of evidence-based medicine, cost effective health care, patient decision making and public health. SMDM helps you to be more than just a face in the crowd. The connections you make through SMDM can help you build a network of long-term contacts to help you throughout your career.
"SMDM is a professional home, providing an excellent international forum to promote methodologically rigorous, multidisciplinary research for health care decision making, networking and career development."continue »
MDM offers rigorous and systematic approaches to decision making that are designed to improve the health and clinical care of individuals and to assist with health policy development.