Using Synergies Between Survey Statistics and Causal Inference to Improve Generalizability of Clinical Trials

While randomized clinical trials provide a gold standard for causal inference, that inference can be with reference to a population that does not match the actual target population for the treatment. Much of the foundational underpinning of causal inference overlaps with survey statistics, as both focus on issues of selection bias and missing data. I will discuss these synergies and how they might be further leveraged to improve the transportability of clinical trial results to real world settings.