Intraclass correlation coefficients for cluster randomized trials in primary care: data from the MRC Trial of the Assessment and Management of Older People in the Community

Abstract

The cluster randomized trial, in which groups rather than individuals are allocated to different interventions, is an increasingly popular design. In cluster trials observations on individuals within the same cluster may be correlated, and this lack of independence must be taken into account when designing a new trial. We present intraclass correlation coefficients derived from the Medical Research Council Trial of the Assessment and Management of Older People in the Community. This is a UK-based randomized trial comparing different methods of multidimensional screening for people aged 75 years and over. One hundred six general practices and over 30,000 individuals are taking part. Estimates of the intraclass correlation coefficients were obtained using one-way analysis of variance. This is by far the broadest collection of intraclass correlation coefficients for older people at the level of the primary care clinic published to date. The intraclass correlation coefficients presented will be useful in calculating sample sizes for cluster randomized trials and surveys at the primary care clinic level. In conjunction with other papers presenting collections of intraclass correlation coefficients, this paper should help to improve the quality of cluster randomized trials and hence help lead to more reliable estimates of the effectiveness of health care interventions.