Crossover period data and populations

Track completion missingness and deviations

Retain participant, assigned sequence, period, actual treatment and assessment timing. Missing periods are not equivalent to independent missing participants.

Flow diagram

Crossover period data and populationsF1.S1Validate sequence and periodrecordsF1.S2Track completion by periodF1.S3Derive analysis variablesF1.S4Define estimand-specificpopulationsF1.S5Freeze analysis-readyspecificationsF1.N1Period data specificationVariableSourceDerivationMissingnessPopulation ruleParticipantSequencePeriodTreatmentOutcomeWashoutF1.N2Assumptions and rationaleF1.N3Method and protocol evidenceF1.S0 Start flowCrossover period data and populationsF1.S999 End flowEnd flow

Steps

Validate sequence and period records

Reconcile randomized sequence, received treatment and assessment dates.

Track completion by period

Keep missing outcomes, withdrawals and washout deviations distinct.

Derive analysis variables

Preserve participant identifiers, period, sequence and treatment indicators.

Define estimand-specific populations

Justify handling of incomplete pairs and protocol deviations.

Freeze analysis-ready specifications

Retain query resolution and blinded review decisions.

Crossover period data and populations

End flow

Tables and supporting information

Period data specification

Period data specificationVariableSourceDerivationMissingnessPopulation ruleParticipantSequencePeriodTreatmentOutcomeWashout

Store definitions here; avoid identifiable participant data.

Assumptions and rationale

Dropping all incomplete sequences without justification can create selection bias and reduce usable information.

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