Putting theory into practice: a cluster randomized trial with a small number of clusters

Many of the difficulties encountered in the design, organization and analys is of cluster randomized trials arise from the dual nature of such trials; that is, they focus on both the cluster and the individual. A trial now in progress to compare three methods of promoting secondary prevention of coro nary heart disease in primary care includes only 21 general practices, but 2142 patients, and thus contains the problems of both small and large sampl es. With only seven practices in each arm, the trial demanded carefully res tricted randomization, may be difficult to analyse, and risks loss of power if one practice should drop out. At the same time, the large number of pat ients makes for an expensive and administratively complex study. The simult aneous demands of clarity and thoroughness point to an analysis at both clu ster and individual level. With two different approaches, however, there ma y be difficulties of presentation, even if the results agree, and additiona l problems of interpretation if they do not. Finally, practical considerati ons may conflict with theoretical demands. Since the trial contained a serv ice element, all patients with heart disease had to be included, even thoug h it would have been more efficient to take only a sample of patients from some practices. Copyright (C) 2001 John Wiley & Sons, Ltd.