Back to Glossary

    Data & Quality

    Estimand

    An estimand is a precise description of the treatment effect a clinical trial aims to estimate, defined by five attributes: treatment, population, variable (endpoint), handling of intercurrent events, and population-level summary.

    The estimand framework, formalized in ICH E9(R1) (2019), forces sponsors to specify before unblinding exactly what question their analysis answers. It clarified ambiguities around how to handle treatment discontinuation, switching, rescue medication, and death; intercurrent events that complicate the relationship between assigned treatment and outcome.

    Five strategies for handling intercurrent events: treatment policy, hypothetical, composite variable, while on treatment, and principal stratum. Each yields a different estimand and a different statistical analysis.

    References

    Related terms

    See how Kitsa applies this in your trials

    Kitsa's agentic AI compresses protocol development, site selection, and patient pre-screening from months into weeks.

    Request a demo