Stepped Wedge Designs for Two Treatments: Power Analysis and Considerations

• Activity Number: 323 - Estimating Treatment Effects: Applications in Health Policy
• Type: Contributed
• Date/Time: Wednesday, August 5, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Health Policy Statistics Section
• Author(s): Phillip Sundin* and Catherine Crespi
• Companies: University of California, Los Angeles and UCLA Department of Biostatistics, Jonsson Comprehensive Cancer Center
• Keywords: stepped wedge design; clinical trial; cluster randomized trial; trial design; power analysis

Abstract

Stepped wedge designs (SWD) have become an increasingly popular form of cluster randomized trials in both public health and clinical research. However, most research has focused on the SWD with one treatment. There is a lack of literature for SWDs with multiple interventions. By implementing two interventions with a combined condition receiving both treatments in a stepped wedge factorial design, there is potential for increased efficiency, a decreased total number of clusters, and estimation of interaction effects, compared to conducting two SWD trials. We present a model for a SWD trial with two treatments and a combined condition and develop methods to calculate power for the treatment and interaction effects. This model is extended to allow for cohort studies and for cluster-by-time random effects. We examine features of designs that impact power and compare power for different design options.