Examining Disease Risk Factors by Mining Publicly Available Information

Human disease and associated risk factors are of great interest in the medical field. The skyrocketing cost of health care makes the understanding of disease risk factors of even greater importance. When risk factors are well understood for a disease it is possible to educate the public to reduce their risk by avoiding risk factors that they can control. There are publicly available data stores which document risk factors facing the general public in the United States. In particular, the Behavioral Risk Factor Surveillance System (BRFSS) has been maintained by the Center for Disease Control (CDC). This study examines the disease of diabetes using a set of risk factors which the BRFSS maintains. A logistic regression model is created which models diabetes as a function of risk factors. Nine years of existing data from BRFSS was used in building this model (for the years 2002 to 2010). The generated model shows promise in modeling diabetes as a function of risk factors and correctly identified obesity as an important risk factor for diabetes. Studies of this nature can be very cost effective and may generate insights into the risks of disease.