Randomized crossover trial
Can the same participants receive multiple interventions in randomized sequences?
When to use this project
Compare reversible intervention effects within participants when the condition is sufficiently stable and period or carryover effects can be addressed.
How it differs
Participants contribute observations under more than one treatment, unlike parallel groups. Randomization assigns sequences. Persistent effects, cure or rapidly changing disease can make crossover unsuitable.
Decisions before you start
- Justify reversibility, washout and the length of each treatment period.
- Prespecify treatment, period and sequence handling and the repeated-measure covariance structure.
- Plan missing periods and discontinuation. Do not choose an analysis retrospectively from a preliminary carryover significance test.
How the flows work together
Each flow below links to its diagram, steps and tables. Flow order organizes the project; it does not imply that all work is strictly sequential. Data assessment, analysis planning and protocol refinement may inform one another.
1. Crossover trial overview
Overview · Question through within-participant trial reporting
Use this overview to state the purpose and link to detailed decisions. Keep a versioned protocol and decision register, and add evidence references for the rationale.
2. Recruitment and sequence allocation
Selection · Assign participants to treatment sequences
Randomize treatment sequences, retaining participant assignment even when periods are missed. Treatment order and masking procedures should be documented independently.
3. Periods treatments and washout
Conduct · Specify period timing and carryover
Period length and washout depend on the intervention and disease, not a universal interval. Address residual treatment effects and changes in the underlying condition before choosing crossover.
4. Endpoints and estimands
Design · 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.
5. Paired design and sample size
Statistics · 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.
6. Period data and populations
Data · Track completion missingness and deviations
Retain participant, assigned sequence, period, actual treatment and assessment timing. Missing periods are not equivalent to independent missing participants.
7. Crossover analysis plan
Statistics · Specify period sequence and sensitivity
Prespecify treatment and period effects and a suitable repeated-measure covariance structure. Avoid selecting a first-period-only analysis on the basis of a preliminary carryover test. Investigate plausible carryover through design and sensitivity analyses.
Equations and statistical interpretation
These explanations describe notation and assumptions. Flow does not calculate these values.
Paired treatment contrast
dᵢ is the treatment A minus B outcome difference for participant i and s_d is the sample standard deviation of those differences.
This simple illustration assumes independent participants with complete pairs and a suitable treatment contrast. Period effects, carryover and missing observations may require a richer prespecified model.
Variance of a paired treatment difference
σ_A and σ_B are the outcome standard deviations under the two treatments and ρ their within-participant correlation.
This variance identity explains why within-person correlation matters. A full crossover calculation must also reflect the planned estimator, period structure and missing observations.
Published example
Williams et al. (2015), PATHWAY-2: a randomized, double-blind, crossover trial
Inspect treatment sequences, repeated measurements and period handling. PATHWAY-2 compared four treatments; the paired-difference illustration below is deliberately simpler.
This publication illustrates a study design, not an endorsement of every method or a reproduction of its results. Follow its protocol, supplement and update history when studying the example.
Hypothetical scenario
A team wants to compare two short-acting symptom-control interventions in adults with a stable chronic condition.
- Crossover trial overview: Assess whether symptoms and treatment effects support crossover, then approve and register the protocol.
- Recruitment and sequence allocation: Randomize concealed treatment sequences and track sequence assignment.
- Periods treatments and washout: Specify period durations, washout, assessments and adherence.
- Endpoints and estimands: Define the within-participant treatment contrast and intercurrent events.
- Paired design and sample size: Calculate precision or power using within-participant variability and incomplete-period scenarios.
- Period data and populations: Retain participant, sequence, period and treatment identifiers with missingness reasons.
- Crossover analysis plan: Fit the prespecified repeated-measure analysis and assess sensitivity to period and carryover assumptions.
Deliver a populated protocol, decision log, data specification, and planned outputs. Enter numerical findings only after analysis and verification. Keep decisions made before seeing results distinguishable from later amendments.
Sources
- CONSORT 2025: reporting randomized trials
- SPIRIT 2025: randomized trial protocols
- CONSORT extension for randomized crossover trials (2019)
- ICH E9(R1): estimands and sensitivity analysis (2019)
- WHO: clinical trials and prospective registration
- ISRCTN: what the clinical study registry records
- ClinicalTrials.gov: about the registry
Source check: 2026-09-17. Record the version of each reporting or appraisal tool used in your actual protocol.
Compare all Clinical trials project types · Open editor in a new tab