The Stepped-Wedge Clinical Trial: Evaluation by Rolling Deployment

INTRODUCTION

This JAMA Guide to Statistics and Methods discusses the stepped-wedge approach to cluster randomized clinical trial design, in which clusters are randomized to the order in which they receive the experimental regimen.

Cluster randomized trials are studies in which groups of individuals, for example those associated with specific clinics, families, or geographical areas, are randomized between an experimental intervention and a control.¹ A stepped-wedge design is a type of cluster design in which the clusters are randomized to the order in which they receive the experimental regimen. All clusters begin the study with the control intervention, and by the end of the trial (assuming no unexpected and unacceptable safety issues arise), all clusters are receiving the experimental regimen.

USE OF THE METHOD

Why Is a Stepped-Wedge Clinical Trial Design Used?

Cluster randomized trials have been performed for many decades, even centuries,² but the statistical underpinnings of such designs have been worked out only relatively recently.^3,4 The primary motivation for a cluster design is to study treatments that can be delivered only in a group setting (eg, an educational approach in a classroom setting) or to avoid contamination in the delivery of each regimen (eg, a behavioral intervention that could be delivered individually but in settings in which those randomized to different approaches are in close contact with each other and might learn about and then adopt the alternative regimen).¹ Clusters are typically identified prospectively and randomized to receive the experimental or control intervention. However, there are exceptions, such as the ring vaccination trial conducted during the 2014-2015 Ebola epidemic, in which clusters were defined around newly identified cases.⁵

If a cluster randomized trial is deemed necessary or desirable in a specific setting, but resource limitations permit only a gradual implementation of the experimental regimen, a stepped-wedge design may be considered as the fairest way to determine which clusters receive the experimental regimen earlier and which later. Stepped-wedge designs have benefits similar to those of crossover trials because outcomes within a cluster may be compared between the time intervals in which a cluster received the control and the experimental interventions. This controls for the unique characteristics of the cluster when making the treatment comparison. One attractive aspect of stepped-wedge designs is that all participants in all clusters ultimately receive the experimental regimen, thereby ensuring that all participants have an opportunity to potentially benefit from the intervention. This can be advantageous when strong beliefs exist regarding the efficacy of a treatment regimen. When limited resources preclude making the treatment regimen widely available from the start, the use of randomization to determine which clusters get early access to the treatment regimen may appeal to participants’ sense of fairness.

Description of the Stepped-Wedge Clinical Trial Design

Important considerations in designing a stepped-wedge trial ...