(036) The ICH E9(R1) Estimand Framework in the Context of Real-World Data and Observational Studies

Background: Despite growing interest in Real-World Data (RWD) from regulatory agencies, health technology assessors and payers, concerns over the credibility and relevance of observational studies persist. Transparency issues in communicating key study attributes hinder proper assessment and reduce these studies´ utility in healthcare decision making. While the Target Trial Emulation (TTE) and PICO frameworks are well known to epidemiologists, regulators are more familiar with the estimand framework introduced by the ICH E9(R1) addendum. Therefore, bridges between these "languages" are needed to clearly communicate the clinical question of interest, as different stakeholders have different objectives.

Objectives: To introduce the ICH E9(R1) estimand framework

Methods: We aim to introduce the ICH E9(R1) estimand framework, highlight its practical implementation in observational studies using RWD, and explore its relationship with other frameworks such as PICO and the TTE approach.

Results: The estimand framework comprises five attributes: target population, treatment condition, endpoint, population level summary, and strategies for handling intercurrent events. Like PICO, it is used to specify the study objective and research question. Unlike PICO, it also places emphasis on the summary measure and the definition and handling of intercurrent events occurring during follow-up, e.g. treatment discontinuation, switching or additional medication. Moreover, the estimand framework clearly differentiates the "what" from the "how", a distinction less evident in the TTE framework.

Conclusions: The estimands framework assists researchers in clearly defining scientific questions. It strengthens communication across stakeholders about key study attributes, and helps to align study objective, design, conduct, statistical analysis, and interpretation. It complements the PICO and TTE approaches.