Planning stepped wedge cluster randomized trials to detect treatment effect heterogeneity

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Epidemiology

Background:

• Stepped wedge designs are increasingly used for intervention evaluation.
• Existing analytical methods primarily focus on average treatment effects.
• There is a need for methods to evaluate treatment effect heterogeneity in stepped wedge trials.

Purpose of the Study:

• To derive variance formulas for analyzing treatment effect heterogeneity in stepped wedge designs.
• To provide methods for sample size and design configuration for assessing both average and heterogeneous treatment effects.
• To offer a framework for efficient average treatment effect analyses using covariate adjustment.

Main Methods:

• Development of novel variance formulas for confirmatory analyses of treatment effect heterogeneity.
• Application of the framework to both cross-sectional and closed-cohort stepped wedge designs.
• Utilizing a simulation study and real-world trial data (Lumbar Imaging with Reporting of Epidemiology Trial) for validation.

Main Results:

• Novel variance formulas for treatment effect heterogeneity are derived.
• The framework enables efficient average treatment effect analysis via covariate adjustment.
• Optimal cluster allocations are identified to maximize precision for both average and heterogeneous effects.

Conclusions:

• The developed methods fill a critical gap in stepped wedge trial design and analysis.
• These methods support rigorous evaluation of intervention effects across patient subpopulations.
• The findings enhance the precision and efficiency of stepped wedge trial planning and execution.