Power calculation in stepped-wedge cluster randomized trial with reduced intervention sustainability effect

Area of Science:

• Clinical Trials Methodology
• Health Services Research
• Biostatistics

Background:

• Stepped-wedge designs are increasingly used in pragmatic clinical trials for health services research.
• These designs are favored when interventions are known to be effective and safe, or due to logistical/financial constraints necessitating sequential implementation.
• Existing literature lacks methods for stepped-wedge trials with distinct active implementation and sustainability phases where intervention effects may diminish.

Purpose of the Study:

• To address the gap in power analysis for stepped-wedge cluster randomized trials experiencing a reduced intervention effect during a sustainability phase.
• To develop an analytical formula for power calculations tailored to this specific trial scenario.

Main Methods:

• The study focuses on developing a novel analytical formula for power analysis.
• This formula is designed to accommodate stepped-wedge designs with a two-part study period (active implementation and sustainability phase).

Main Results:

• An analytical formula for power analysis has been derived for stepped-wedge designs with a sustainability phase.
• This formula aids in the design of cluster randomized trials where intervention effects may wane over time.

Conclusions:

• The developed formula provides a crucial tool for researchers designing stepped-wedge trials in health services, particularly those with extended study periods and potential for effect modification.
• This methodology supports more accurate power estimations, enhancing the rigor of pragmatic clinical trials evaluating interventions in real-world settings.