Beyond the Dose: What is the Estimand Regimen? Explained as a Clinical Trial Framework

October 11, 2025 •

3 min read

Since its formal adoption in 2019 via the ICH E9(R1) guideline addendum, the estimand framework has become central to clinical trial design. It is important to clarify that "Estimand regimen" is a misconception; an estimand is not a drug protocol but a precise statistical framework for defining the treatment effect a trial aims to measure.

Quick Summary

The estimand framework is a structured approach used in clinical trials to clearly define the treatment effect of interest, particularly how to account for unexpected events like treatment discontinuation or rescue medication use. It aligns trial objectives with design and analysis to improve transparency and interpretability of results.

Key Points

Misconception Clarified: The term "Estimand regimen" is incorrect; an estimand is a statistical framework, not a drug regimen.
Structured Framework: The Estimand Framework, developed by ICH, provides a precise, five-attribute structure for defining the treatment effect in a clinical trial.
Intercurrent Events (ICEs): A key focus is defining how to handle post-randomization events like discontinuation or rescue medication use that can complicate results.
Multiple Strategies: There are five main estimand strategies (Treatment Policy, Hypothetical, Composite, Principal Stratum, While on Treatment) for addressing ICEs, each answering a different clinical question.
Enhanced Transparency: The framework improves the clarity and transparency of clinical trials, aligning objectives with design and analysis for more reliable and interpretable results.
Informs Stakeholders: Estimaands help various stakeholders, from regulators to patients, better understand the reported treatment effects and their real-world relevance.

What is the Estimand Framework?

The estimand framework, formalized by the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) E9(R1) addendum in 2019, provides a structured approach for precisely defining the objective of a clinical trial. It specifies what treatment effect is being measured, which is especially important because clinical trials often encounter post-randomization events that can complicate interpretation. These are known as intercurrent events (ICEs). The framework enhances clarity and transparency by requiring researchers to pre-specify how these events will be handled in the analysis.

The Five Attributes of an Estimand

Defining an estimand involves specifying five key attributes:

Treatment: The specific intervention being compared.
Population: The target patient group.
Variable (Endpoint): The outcome measure used.
Handling of Intercurrent Events (ICEs): How post-randomization events are addressed.
Population-level Summary: How individual data are summarized for comparison.

Strategies for Handling Intercurrent Events

The estimand framework outlines five main strategies for handling ICEs, each addressing a different clinical question:

Treatment Policy Strategy: Evaluates the effect of being assigned a treatment, regardless of whether ICEs occur. All data are included to reflect real-world outcomes.
Hypothetical Strategy: Estimates the treatment effect in a hypothetical scenario where the ICE did not happen, helping to understand the treatment's effect without confounding events.
Composite Strategy: Incorporates the ICE directly into the endpoint definition, creating a combined variable where the ICE often indicates a negative outcome.
Principal Stratum Strategy: Focuses the analysis on a subgroup of patients in whom a specific ICE would not occur under any treatment.
While on Treatment Strategy: Analyzes the treatment effect only up to the point that an ICE occurs, censoring data collected afterward.

Comparing Estimand Strategies in Action


Feature	Treatment Policy	Hypothetical	Composite	While on Treatment
Focus	Pragmatic, real-world effect of treatment assignment.	Efficacy of treatment had patients adhered to protocol.	Treatment success incorporating an ICE as a negative outcome.	Effect of treatment only while receiving the intervention.
Analysis Data	All patient data included regardless of ICEs.	Estimates based on a hypothetical scenario where ICEs did not occur.	Data integrated into a combined endpoint that includes ICEs.	Data collected only up to the point of the ICE.
ICE Handling	Ignored in the analysis; included in the overall effect.	Imputed or modeled away to estimate effect without ICE.	Woven into the endpoint definition (e.g., treatment failure).	All data post-ICE are censored or excluded.
Example	Analyzing HbA1c change including data from patients who took rescue medication.	Estimating HbA1c change as if patients had never taken rescue medication.	Defining treatment failure as either not achieving HbA1c target or taking rescue medication.	Analyzing liver biopsy results only until a patient discontinues the study drug.

Benefits of Using the Estimand Framework in Pharmacology

Implementing the estimand framework in pharmacology brings several benefits:

Increased Clarity: Defines trial objectives transparently from the start.
Enhanced Alignment: Ensures the clinical question, trial design, and analysis are consistent.
Improved Transparency: Reduces bias and clarifies communication of findings to various stakeholders.
Better-Informed Decisions: Allows different stakeholders to use trial results relevant to their specific viewpoints.
Robustness of Results: Prompts sensitivity analyses to test conclusions under different ICE assumptions.

Conclusion

The term "Estimand regimen" is a misunderstanding; the Estimand Framework is a crucial statistical approach for clearly defining the target of estimation in clinical trials, not a treatment protocol. By detailing the treatment, population, variable, summary, and ICE handling, it aligns trial components and improves the transparency and reliability of pharmacological research. This leads to better-informed decisions in drug development and clinical practice.

For more information on the official guidelines, consult the full text of the {Link: FDA https://www.fda.gov/regulatory-information/search-fda-guidance-documents/e9r1-statistical-principles-clinical-trials-addendum-estimands-and-sensitivity-analysis-clinical}.

Frequently Asked Questions

An estimand is a precise description of the treatment effect to be estimated in a clinical trial. It outlines the specific clinical question being asked and defines the treatment, population, endpoint, summary measure, and handling of intercurrent events to provide clarity.

ICEs are events that occur after a participant has started a clinical trial that can affect the interpretation or existence of outcome measurements. Examples include a patient discontinuing treatment, using a rescue medication, or switching to an alternative therapy.

While an ITT analysis can correspond to a 'treatment policy' estimand strategy, it doesn't provide the same level of granular detail. The estimand framework clarifies which events are considered intercurrent and explicitly defines the strategy for handling them, which an ITT analysis alone does not.

Yes, a clinical trial can and often does define multiple estimands. This allows for addressing several different clinical questions and provides a more complete picture of the treatment effect for various stakeholders.

The framework is crucial for pharmacology as it improves the transparency, reproducibility, and interpretation of clinical trial results. It ensures that the effects of a drug, particularly in the face of real-world complexities like non-adherence or rescue medication use, are clearly communicated to regulators and clinicians.

The hypothetical strategy involves defining a scenario where a specific intercurrent event would not occur. It is used to estimate the treatment effect under idealized conditions, assuming perfect adherence or no confounding events.

In the composite strategy, an intercurrent event is included as part of the overall endpoint definition. This is useful when the ICE is informative about a poor outcome, such as incorporating the need for surgery as a treatment failure in a clinical trial for nasal polyps.