Cluster Unit Randomized Trials

2. Introduction

Cluster randomization trials (CRTs) are experimental studies in which intact social units (clusters), such as families, schools or even entire cities, rather than independent individuals, are randomly allocated to intervention groups. The purpose of this chapter is to provide an introduction to the basic principles that apply to the design and analysis of CRTs, using examples throughout to illustrate the key points.

Although CRTs were seen only infrequently before the mid-1980s, their popularity has increased dramatically over the last 20 years, particularly in the evaluation of innovations in health care. Figure 1, adapted from Bland (2004), shows that the growth in CRTs from the mid-1990s has been particularly rapid. Perhaps not surprisingly, the methodological foundation for CRTs has been much slower to develop, with the first texts dealing exclusively with this design only appearing in the last 10 years (Murray, 1998; Donner and Klar, 2000). A review of more recent developments may be found in Campbell et al., (2007).

Figure 1.

Cluster Randomization Trials Published 1981-2003

Graph showing the growth of cluster randomization trials published from 1981 to 2003.

Source: Bland J.M. (2004) “Cluster randomised trials in the medical literature:Two bibliometric surveys.” BMC Medical Research Methodology;4: Figure 1, p: 4.

Bland J.M. (2004). Cluster randomised trials in the medical literature:Two bibliometric surveys. BMC Medical Research Methodology, 4: 1-6.
Murray D.M. (1998). Design and analysis of group-randomized trials. New York: Oxford University Press.
Donner A., Klar N. (2000). Design and analysis of cluster randomization trials in health research. New York: Oxford University Press.