Kulldorff's Scan method (Kulldorff and Nagarwalla 1995, Kulldorff 1997) can detect local space or space-time clusters in group-level data.
The scan statistic uses a circular or cylindrical window to identify excesses of cases in space and time. At each spatial zone, a circular window increases in size until it reaches an upper size limit. At each spatio-temporal location, a cylindrical window increases in size in both space and time until it reaches an upper size limit.
The scan statistic provides a measure of whether the observed number of cases is unlikely for a window of that size, using reference values from the entire study area. By searching for clusters without specifying their size or location, the method avoids pre-selection bias.
Kulldorff (1997) developed two models, a Poisson model and a Bernoulli model. For a small number of cases, the two models are similar. The Bernoulli model is best for questions about case and control samples, while the Poisson model better answers questions with case and population-at-risk counts. At this point, ClusterSeer implements the Poisson method.
The scan statistic has been applied to childhood leukemia in Sweden (Hjalmars 1996) and upstate New York (Kulldorff and Nagarwalla 1995) and to breast cancer in the northeastern United States (Kulldorff et al. 1997).
