Abstract
A Real-World Evidence (RWE) scientific working group of the American Statistical Association Biopharmaceutical Section endeavors to elevate statistical practices of generating real-world evidence to support regulatory decision making. Randomized clinical trials have been the gold standard for evaluating the safety and efficacy of medical products. However, in many cases, their costs, duration, limited generalizability, and ethical or technical infeasibility might have motivated some to consider alternatives, such as real-world studies. Yet, the studies conducted in a real-world setting may be susceptible to the lack of randomization and the presence of confounding bias. To generate robust real-world evidence (RWE) from analyzing real-world data (RWD), we have proposed an RWE Causal Inference Roadmap with targeted learning. Three key steps in the roadmap are (1) define a target estimand that aligns with the causal estimand of the study objective; (2) select an efficient estimator for estimating the target estimand and an estimator of its uncertainty; and (3) evaluate the robustness of conclusions to violations of untestable causal assumptions. In this article, we demonstrate how to apply the roadmap to generate RWE using synthetic data based on a real oncology trial.