Introduction and Overview

• Published in: Statistical Detection and Surveillance of Geographic Clusters pp. 25 - 44
• Main Authors: Rogerson, Peter, Yamada, Ikuho
• Format: Book Chapter
• Language: English
• Published: Chapman and Hall/CRC 2009
• Subjects: Cancer Mortality; Cancer Mortality Data; Areal Data; Confirmatory Methods; Zip Code Zone; LBW Rate; Luc Anselin; Standardized Mortality Ratio; GIS Environment; Information URL; Sids Rate; Disease Cases; Spatial Clustering; CMF; ICD-9 Code; Sids; Spatial Dependence; Space Time Scan Statistics; Case Control Data; Prostate Cancer Deaths; Breast Cancer Mortality; Simple Random Assignment; Exploratory Spatial Data Analysis Techniques; Prostate Cancer; Sudden Infant Death Syndrome
• DOI: 10.1201/9781584889366-5

Imagine yourself in the role of a public health official in an urban area. You are asked to monitor data on admissions to emergency departments for respiratory distress and to report on any abnormal increases that may occur. As part of your response, you decide to both tabulate and map the daily, weekly, and monthly reports of such occurrences. As you examine your first map depicting the location of new cases, how do you decide whether the map of occurrences has an interesting pattern? There may be apparent clusters of incidents, but these may simply reflect geographic patterns of population density. Perhaps you have accounted for population density and also for the age structure of the population. Is there still a pattern on the map that deviates significantly from some simple random assignment of cases to the population? The geographic pattern may reflect other factors-perhaps individuals living near hospitals are more likely to be admitted for respiratory distress.