The Analysis of Asthma and Exposure Data using Geographic Information Systems and Data Mining Information

Data mining was used to investigate two facts: that African Americans represent 15% of the population but 38% of patients needing treatment in a hospital emergency room for shortness of breath and that African Americans have extremely high rates of lung cancer while reporting lower rates of cigarette smoke and exposure to toxic chemicals. Geographic Information Systems (GIS) containing locations of businesses that release toxic chemicals were combined with information concerning patient treatment and outcomes for shortness of breath. It was determined that African Americans routinely under-report exposure, and this under-reporting can result in reduced treatment options. Introduction This paper discusses the statistical methodology needed to combine environmental data with medical data by providing a specific example to investigate the relationship between environmental factors and the need for treatment of lung problems. Data collected in three clinical studies were examined to determine patient needs and patient screening. The environmental information is stored in a Geographic Information System (GIS). The GIS can create maps linking medical data to environmental location to examine patterns and relationships in the data. To properly analyze the data, data mining tools must be used. Multiple levels of environmental factors combined with multiple patient factors add a level of complexity to the data that cannot be examined using standard statistical methods. Only data mining tools, designed to work with large, complex databases can examine the combined datasets.