swdpwr: A SAS macro and an R package for power calculations in stepped wedge cluster randomized trials

Area of Science:

• Biostatistics
• Clinical Trials Methodology
• Public Health Research

Background:

• Stepped wedge cluster randomized trials (SWDs) are increasingly utilized in public health and clinical research.
• Existing power calculation methods for SWDs often rely on normal approximations for binary outcomes, which may lack accuracy.
• Newly developed methods for binary outcomes in SWDs lack accessible software implementation.

Purpose of the Study:

• To present user-friendly software for power calculations in stepped wedge cluster randomized trials (SWDs).
• To accommodate various settings, including continuous and binary outcomes, different designs, and correlation structures.
• To bridge the gap between advanced statistical methods and practical application in SWD design.

Main Methods:

• Development of a SAS macro (%swdpwr), an R package (swdpwr), and a Shiny application for power calculations.
• The software supports cross-sectional and cohort designs, binary and continuous outcomes, and various correlation structures (exchangeable, nested exchangeable, block exchangeable).
• Accommodates marginal and conditional models, different link functions, time effects, and unequal cluster numbers per sequence.

Main Results:

• The developed software, swdpwr, offers an efficient tool for designing and analyzing SWDs.
• It addresses the need for more accurate power calculations by implementing recent methodological advancements.
• Provides accessible power calculation tools for investigators conducting SWDs.

Conclusions:

• User-friendly software (SAS macro, R package, Shiny app) has been developed for power calculations in SWDs.
• The software makes computationally efficient, non-simulation-based power methods accessible for continuous and binary outcomes.
• Caters to diverse user needs, including those without SAS/R expertise via the online Shiny app.