ABSTRACT
The spatial analysis of small area health data is a topic which has developed greatly in the last decade. In this development, there has been considerable interest in the analysis of clustering of disease. This has come about in response to a need within public health to be able to analyse disease maps with a view to establishing whether clustering exists, particularly where a possible environmental hazard may be related to the adverse disease outcome. Much of the methodology developed in this area has focussed on hypothesis testing and little has focussed on statistical modelling of clustering. Recent reviews of these developments can be found in Lawson and Kulldorff (1999), and Lawson (2001), Chapter 6. There are many advantages to modelling of clusters, not least of which is the flexibility to introduce covariates and to be able to include a range of descriptive parameters within the fitting process. While relatively little development of spatial models has been witnessed, the analysis of clustering in spatiotemporal small area health data has seen even less development. In Chapter 1 of this volume, an introduction of the concepts of cluster
modelling are presented.
