From Data to Evidence: A Framework for RWD Study Design & Causal Inference

Real-world data (RWD) is increasingly used for causal inference in healthcare research, but generating credible, decision-ready insights requires more than access to data. It demands intentional alignment between the causal question, the study design and the data source. Too often, organizations treat data acquisition as the finish line, when in reality even large, well-curated datasets can produce misleading results if the methods are poorly chosen or the data is mismatched to the question being asked.

This webinar presents a practical framework for designing rigorous RWD studies for causal inference. A foundational premise is that study design and the selection of analytical methods cannot be predetermined in the absence of information about the data source. But starting with the data and working backward to the question is also a common error that even experienced teams make under time and resource pressure.

Attendees will learn how to transparently manage the interplay between question and data using the target trial framework: beginning with a clearly framed research question based on a Target Trial and iteratively adapting it until either finding a viable study design that maps to the available data, or concluding that the data is not fit-for-purpose.

The seminar will focus on two critical aspects of conducting an RWD study: 1) starting with a strong and structured framework to ask a clear question and study design, and 2) a structured approach for assessing fitness-for-purpose within the context of the study question and study design. Attendees will leave with practical knowledge and a set of decision points they can apply immediately to their own work.

Speakers

Jeffrey S. Brown, PhD, Chief Scientific Officer, TriNetX

Miguel Hernán, MD, PhD, ScD, Director of CAUSALab, Professor of Biostatistics and Epidemiology, Harvard