The attributable estimand: A new approach to account for intercurrent events

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Pharmacoeconomics

Background:

• Intercurrent events complicate treatment effect estimation in clinical trials.
• Accurate calculation of treatment effects requires robust methods for handling these events.

Purpose of the Study:

• To propose and evaluate an attributable estimand strategy for addressing intercurrent events.
• To compare this strategy with existing methods like hypothetical and treatment policy estimands.

Main Methods:

• Developed a composite estimand strategy for treatment-related intercurrent events (e.g., adverse events, lack of efficacy).
• Utilized a hypothetical estimand strategy for unrelated intercurrent events.
• Conducted simulation studies inspired by chronic obstructive pulmonary disease (COPD) trials.
• Analyzed data from a completed COPD trial to illustrate the approach.

Main Results:

• The attributable estimand strategy provides a nuanced approach to handling intercurrent events.
• Simulation results demonstrate the performance of the proposed strategy compared to alternatives.
• The analysis of the COPD trial showcases practical application and interpretation.

Conclusions:

• The attributable estimand strategy offers a valuable tool for improving the precision of treatment effect estimation in the presence of intercurrent events.
• This method enhances the interpretability of trial results, especially in complex disease areas like COPD.