Crossover endpoints and estimands

Define period outcomes and effects

Define when an endpoint is measured in each period and what within-person contrast it supports. Treatment discontinuation and missing periods may change both interpretability and the analysis population.

Flow diagram

Crossover endpoints and estimandsF1.S1Define period outcomesF1.S2Define treatment contrastsF1.S3Handle period and sequence contextF1.S4Specify intercurrent-event handlingF1.S5Prioritize endpointsF1.N1Crossover endpoint registerEndpointPeriod timingContrastIntercurrenteventsPriorityPrimarySecondarySafetyF1.N2Assumptions and rationaleF1.N3Method and protocol evidenceF1.S0 Start flowCrossover endpoints and estimandsF1.S999 End flowEnd flow

Steps

Define period outcomes

Specify measurement timing and identical assessment procedures across treatments.

Define treatment contrasts

State the within-participant effect and target population.

Handle period and sequence context

Distinguish treatment effect from time and order effects.

Specify intercurrent-event handling

Address withdrawal, rescue treatment and incomplete sequences.

Prioritize endpoints

Record primary, secondary and safety objectives with multiplicity rules.

Crossover endpoints and estimands

End flow

Tables and supporting information

Crossover endpoint register

Crossover endpoint registerEndpointPeriod timingContrastIntercurrenteventsPriorityPrimarySecondarySafety

Match the estimand to period-specific observations.

Assumptions and rationale

A complete-case paired analysis changes the population when incomplete periods are informative.

Method and protocol evidence

Attach the protocol, methodological sources and versioned software specification supporting these decisions. Record citations and provenance; do not enter participant identifiers.

Sources