The Behavioral Risk Factor Surveillance System (BRFSS) is described on the by the CDC as “The BRFSS objective is to collect uniform, state-specific data on preventive health practices and risk behaviors that are linked to chronic diseases, injuries, and preventable infectious diseases that affect the adult population. (…) For most states, BRFSS is the only source of state-based health risk behavior data related to chronic disease prevalence. (…) Currently, BRFSS collects data in all 50 states as well as the District of Columbia and three U.S. territories. BRFSS completes more than 400,000 adult interviews each year, making it the largest continuously conducted multi-mode (mail, landline phone, and cell phone) health survey system in the world”. For that matter, from a data scientist, statistician, or any researcher point of view, the 2013 sample has an enormous amount of information, not to mention the gigantic datafile with almost 500.000 observations of 330 variables. The wealth of data is only matched by the care that must be taken manipulating it. One issue, for example, is the overwhelmingly large number of NAs, caused mostly by the conditional multi-sections of the survey. Some variables show great discrepancy, and values often transmute to categories (e.g. the height related variables, some of them appearing in this paper). Anyway, it is understood that any hassle is the direct consequence of the nature of the survey, it being large, complex and comprehensive. The health and socioeconomic data here present many, almost infinite possibilities of correlation, as seen in the following sections.

This article applies Multidimensional Scaling Techniques on data obtained by applying optimization techniques and differential equations solving onto data from the epidemiologic evolution of the COVID-19 pandemic in Brazil. The MDS provides a visual insight on the how the Brazilian states are performing more closely regarding the epidemiological parameters obtained.