The Estimand Framework and Causal Inference: Complementary Not Competing Paradigms.

Area of Science:

• Biostatistics
• Clinical Trial Design
• Epidemiology

Background:

• The International Council for Harmonisation E9 (R1) guideline introduced an estimands framework for precise treatment effect specification in clinical trials.
• The relationship between the ICH E9 (R1) estimands framework and causal inference remains unclear, despite both defining estimands.

Purpose of the Study:

• To compare and contrast the ICH E9 (R1) estimands framework with causal inference.
• To illustrate how both frameworks can define population-based treatment effects.
• To highlight the complementary nature of these two paradigms in clinical trial methodology.

Main Methods:

• Illustrative examples were used to compare the ICH E9 (R1) estimands framework and causal inference.
• Similarities and differences in defining estimands were analyzed.
• The accessibility and mathematical precision of each framework were discussed.

Main Results:

• Both ICH E9 (R1) and causal inference can define population-based treatment effects.
• The ICH E9 (R1) framework provides a structured, accessible approach for communication.
• Causal inference offers mathematical precision and explicit assumption articulation via tools like causal graphs.

Conclusions:

• The ICH E9 (R1) estimands framework and causal inference are complementary, not competing.
• Integrating both approaches improves the clarity and robustness of clinical trial communication.
• Appreciating concepts from both frameworks strengthens clinical trial design, analysis, and interpretation.