UC Berkeley School of Public Health recently highlighted the Center for Targeted Machine Learning and Causal Inference (CTML) and its work to accelerate the development and evaluation of new drugs using real-world health data. Co-directed by CTML faculty members Maya Petersen, MD, PhD, Mark van der Laan, PhD, and Alan Hubbard, PhD, the Center is advancing statistical methods that bring together causal inference, machine learning, and statistical theory to generate rigorous evidence from observational and clinical trial data. CTML researchers have worked with the U.S. FDA for more than a decade to explore how these methods can support more efficient approaches to drug safety, development, and regulatory decision-making.
Through initiatives including the Joint Initiative for Causal Inference (JICI) and collaborations with the FDA, Novo Nordisk, Gilead Sciences, and academic partners around the world, CTML researchers are developing methods to integrate real-world and randomized clinical trial data, strengthen causal conclusions, and improve the analysis and design of clinical studies. This work has the potential to reduce the time and cost required to evaluate new therapies while maintaining rigorous standards of evidence, ultimately helping bring effective treatments to patients sooner.