The stepped wedge cluster randomised controlled trial is a relatively new study design that is increasing in pop-ularity. It is an alternative to parallel cluster trial designs, which are commonly used for the evaluation of service delivery or policy interventions delivered at the level of the cluster. The design includes an initial period in which no clusters are exposed to the intervention. Sub-sequently, at regular intervals (the " steps ") one cluster (or a group of clusters) is randomised to cross from the control to the intervention under evaluation. This pro-cess continues until all clusters have crossed over to be exposed to the intervention. At the end of the study there will be a period when all clusters are exposed. Data col-lection continues throughout the study, so that each cluster contributes observations under both control and intervention observation periods. It is a pragmatic study design, giving great potential for robust scientific evalu-ations that might otherwise not be possible. Brief history of the stepped wedge cluster randomised trial The stepped wedge cluster randomised trial has been used across several settings for some years, but early stepped wedge designs were sometimes described in other terms such as " waiting list designs " or " phased implementations. " The Gambia hepatitis intervention study (example 1) is probably the earliest and most widely known stepped wedge study. 1 Two systematic reviews, determining the number and breadth of stepped wedge studies, have recently been conducted. 2 3 These reviews reveal that the use of this study design is on the increase and that areas of use are diverse and include HIV, cancers, healthcare associ-ated infections, social policy, and criminal justice. In 2007 Hussey and Hughes 4 first described methods to determine statistical power available when using a stepped wedge design. However, there is a dearth of lit-erature on the more general methodological aspects, such as the rationale for, and conduct of, stepped wedge studies. In this article we illustrate how this new study design differs from the conventional parallel design and its variations. We also give several examples and consider several design and methodological issues, including rationale, sample size, and efficiency com-pared with competing designs, and highlight some important reporting and analysis considerations.
Hemming, K., Haines, T. P., Chilton, P. J., Girling, A. J., & Lilford, R. J. (2015). The stepped wedge cluster randomised trial: Rationale, design, analysis, and reporting. BMJ (Online), 350. https://doi.org/10.1136/bmj.h391