March 2013

Stepped-wedge cluster randomised controlled trials - some variations in design

Karla Hemming and Alan Girling, University of Birmingham

Venue: Room BB016, Brook Building, University of Central Lancashire

Date: 21-03-2013


Stepped wedge cluster randomised trials (SW-CRTs) are being used with increasing frequency in health service evaluation. Conventionally, these studies are designed with equally spaced steps, with an equal number of clusters randomised at each step, and data collected at each and every step. When designing a SW-CRT there are many variations of design patterns with many designs not conforming to the typical stepped wedge study (with clusters randomised at every step and data captured at every step). Here we consider several variations on the design of SW-CRTs and consider implications for power and bias. We introduce the incomplete stepped wedge design where the number of clusters varies at each step, or where at some steps data are not collected. We also show that the parallel CRT with staggered, but balanced randomisation, can be considered a special case of the incomplete SW-CRT, and also consider the implications of individual participant recruitment on selection biases. Finally, we explore extending power to designs in which there are multiple layers of clustering such as wards within a hospital. These issues will be considered through a series of examples, where we consider the implications of these design aspects on power achievable.