Analyzing the Correlations between the Uninsured and Diabetes Prevalence Rates in Geographic Regions in the United States

The increasing prevalence of diagnosed diabetes has drawn attentions of researchers in recently years. Research has been done in finding the correlations between diabetes prevalence with socioeconomic factors, obesity, social behaviors and so on. Since 2010, diabetes preventive services have been covered under health insurance plans in order to reduce diabetes burden and control the increasing of diabetes prevalence. In this study, a hierarchical clustering model is proposed by using Expectation-Maximization algorithm to investigate the correlations between the uninsured and diabetes prevalence rates in 3142 counties in United States for years from 2009 to 2013. The results identified geographic disparities in the uninsured and diabetes prevalence rates of individual years and over consecutive years.