SOLVING THE CLUSTER PUZZLE - CLUES TO FOLLOW AND PITFALLS TO AVOID

Dozens of methods have been proposed for the identification of disease clusters, although only a few are used routinely in published investi gations. New methods, although designed to exploit some particular asp ect of the data or use some specific statistical tool, are rarely comp ared thoroughly in terms of power or performance. Users, when confront ed with the multitude of methods available, often select methods arbit rarily, basing choices on software availability, ease of implementatio n, or use experience rather than considerations of statistical power, possible alternative hypotheses (that is, cluster structure) and likel y confounding. In this review, we extend our typology of disease clust ering methods and apply it to many of the extant methods identifying s trengths, weaknesses and unique features of the methods. We conclude w ith recommendations for which methods should be applied to which types of situations.