ABSTRACT
In these times of increased public awareness of and concern toward sources of environmental pollution and their potential links to disease, reports of a cluster or hot spot of disease in a particular area are commonplace. These concerns have no doubt been heightened, in the United Kingdom at least, by public health issues such as the so-called childhood leukemia cluster around the Sellafield nuclear reprocessing plant in northwest England (Gardner, 1993), over bovine spongiform encephalopathy (popularly known as mad cow disease), and more recently, the foot and mouth disease outbreak in 2001. Much of the controversy surrounding clustering lies in the difficulty in providing an adequate definition. To remove any potential for confusion, we draw the reader’s attention to the distinction between the identification of disease clusters, the topic of this chapter, and the entirely separate statistical technique of cluster analysis, which aims to aggregate variables with similar characteristics in a dataset together to simplify subsequent analysis.
