Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 246 - Data Science
Type: Contributed
Date/Time: Wednesday, August 11, 2021 : 10:00 AM to 11:50 AM
Sponsor: Section on Statistical Computing
Abstract #319039
Title: Sample size determination for stepped-wedge cluster randomized trials with imbalance enrollment using a Shiny app
Author(s): Zhuopei Hu* and Ruofei Du and Songthip T Ounpraseuth
Companies: University or Arkansas for Medical Sciences and University or Arkansas for Medical Sciences and University or Arkansas for Medical Sciences
Keywords: Stepped-wedge cluster randomized trials; Shiny app; sample size; imbalance enrollment
Abstract:

Stepped-wedge cluster randomized trials (SW-CRTs) are pragmatic study designs which have been broadly used in recent years. For standard cross-sectional CRTs or parallel CRTs, the Hussey and Hughes’s (HH) method is widely employed to calculate sample size and power. There exist tools or program packages for determining statistical powers for these traditional CRTs. However, experienced with our current pediatric clinical trial, not all sites were able to transit from a standard care to the intervention or achieve the targeted enrollment number at each period. We sought to examine how much impact the design modification and the extreme enrollment imbalance have on the study power. However, there were no available tools to re-calculate the power based on the observed real enrollment. In this project, we developed an R Shiny web app to examine the effects of incomplete enrollment on the statistical power of the stepped-wedge cluster randomized trial. In particular, we employed a simulated based approach to examine the effect from real-time enrollment, where some sites had incomplete enrollment and other sites had high volume enrollment, in various periods on the power.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program