How Regulators Evaluate Evidence in a Changing Data Landscape
As clinical research evolves, causal inference is playing an increasingly important role in regulatory decision-making. With this shift comes challenges, particularly when balancing scientific rigor with real-world applicability.
A key takeaway from our expert discussion: causal inference should be incorporated from the very beginning of a study. Starting with a well-defined research question leads to clearly outlining the goal of a study. If that goal is causal inference, certain tools can be leveraged such as directed acyclic graphs (DAGs). A DAG can influence various aspects of a study, including::
- Data source selection
- Confounding adjustment strategies
- Sensitivity analyses
- Overall study design
By integrating causal thinking from the beginning, teams can ensure alignment between their objectives and methodologies, something regulators increasingly expect.
The Internal vs External Validity Trade-Off
One of the most complex challenges regulators face is balancing:
- Internal validity (the rigor and reliability of causal conclusions)
- External validity (how applicable findings are in the real world)
There is no one-size-fits-all approach. Regulatory decisions are highly context-dependent, influenced by factors such as disease area, availability of treatments and urgency of unmet need (e.g. rare diseases). Rather than focusing on a single method, regulators evaluate the totality of evidence to determine whether it is sufficient to support decision-making.
The Role of Real-World Evidence
With increasing reliance on real-world data, causal inference methods are becoming more critical. Initiatives across academia, industry, and regulatory bodies, particularly in Europe and North America, are exploring how real-world evidence can inform decisions more effectively. Frameworks and guidance are already beginning to incorporate causal concepts, such as:
- DAGs in methodological recommendations
- Structured approaches like target trial emulation
- Standardized reporting templates
These developments signal a shift toward more formal integration of causal inference in regulatory sciencesettings.
Looking Ahead: What’s Next for Causal Inference?
It is likely causal inference approaches will become even more prominent over the next 3–5 years. Key trends include:
- Increased use of structured frameworks and templates
- Greater collaboration between regulators and researchers
- Potential development of dedicated guidance documents
- Continued innovation in leveraging real-world data
As these practices mature, causal inference may become a standard expectation.
Practical Considerations for Study Design
Regulatory discussions also highlight important considerations in study execution, such as:
- The value of including a concurrent control group, even in augmented RCT designs
- The need to carefully manage the influence of external data sources
- Focusing not just on sample size, but on how data impacts conclusions
Additionally, clinical input remains essential, especially when defining criteria that affect both sample size and real-world feasibility. Causal inference is rapidly reshaping how evidence is generated, evaluated, and used in regulatory settings. While challenges remain, particularly around trade-offs and implementation, the direction is clear: causality is becoming central to modern clinical research and decision-making. For organizations looking to stay ahead, embedding causal frameworks early and aligning with evolving regulatory expectations will be key to success.
