To replicate or not to replicate? Insights and interpretations from a randomized trial duplication initiative

Randomized controlled trials (RCTs) are considered a ‘gold standard’ for evidence generation for several reasons, including their ability to provide causal inferences. However, real-world data (RWD) are increasingly used to complement RCTs for several purposes. These include defining and monitoring the disease landscape, drug development decision-making, informing clinical study design, improving trial efficiency, assessing drug outcomes and supporting regulatory decisions. All uses of real-world evidence (RWE) require analytical rigor, but they do not need to replicate or replace RCTs. The value of RWE should be recognized distinctly from the purposes and research questions of RCTs; indeed, RWE can both complement RCTs and be used to make causal inferences.