ABSTRACT
Most spatial disease surveillance practices deal with disease cluster detection. In the traditional disease cluster detection approach, one first detects the existence of spatial clusters. Once a disease cluster is detected, one needs to trace etiological factors that contribute to the cluster. Alternatively, one might take a nonspatial approach by exploring correlations between disease incidence and environmental exposures over the entire study area. In this chapter, the authors demonstrate both approaches by correlating Parkinson's disease (PD) data and pesticide and herbicide data in Nebraska. They briefly describe how to derive pesticide exposure data. The authors then present a cluster detection approach that links a detected cluster with pesticide exposures, or vice versa, following an ecological study design. They also present a case-control study using PD as cases and multiple sclerosis (MS), Alzheimer's disease, ischemic stroke, and diabetes as separate control groups.
