Crossover sample size and paired precision

Model within-participant variation and precision

Precision depends on within-participant variability and correlation, as well as attrition across periods. Do not substitute an independent-groups calculation for a repeated-measure design.

Flow diagram

Crossover sample size and paired precisionF1.S1Define primary paired contrastF1.S2Specify within-participantvariationF1.S3Account for period structureF1.S4Assess incomplete sequencesF1.S5Document precision or powerscenariosF1.N1Paired design scenariosScenarioDifferenceWithin-personvariationCorrelationAttritionRequired NMain scenarioLowercorrelationIncompleteperiodsF1.N2Variance of a paired treatmentdifferenceParallel-group sample-size assumptions arenot interchangeable with paired crossoverassumptions.F1.N3Method and protocol evidenceF1.S0 Start flowCrossover sample size and pairedprecisionF1.S999 End flowEnd flow

Steps

Define primary paired contrast

Specify outcome scale and testing or estimation objective.

Specify within-participant variation

Record correlation, difference variability and supporting evidence.

Account for period structure

Include planned sequences, treatment periods and analysis model.

Assess incomplete sequences

Plan attrition and missing-period scenarios.

Document precision or power scenarios

Record methods, software and assumptions rather than fixed defaults.

Crossover sample size and paired precision

End flow

Tables and supporting information

Paired design scenarios

Paired design scenariosScenarioDifferenceWithin-personvariationCorrelationAttritionRequired NMain scenarioLowercorrelationIncompleteperiods

Use formulas or simulation compatible with the planned crossover analysis.

Variance of a paired treatment difference

Parallel-group sample-size assumptions are not interchangeable with paired crossover assumptions.

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