Last updated on 01/14/2021
Tang C. (first author). Submitted. “The Intersection of Big Data and Epidemiology for Epidemiologic Research.” In The AMIA 2021 Virtual Clinical Informatics Conference. Online.
The sudden rise of big data in the public health sphere has led to an increased misconception that big data approaches are inherently “garbage in, garbage out” when compared to traditional epidemiological methods. In actuality, big data is comprised of three critical elements: data, technology, and application. Common in big data and epidemiology is a focus on the approaches for solving intricate problems. Perhaps mutually exclusive preferences of the two fields can warrant a tighter integration. For example, epidemiologists are well versed in the science of study design and the art of causal inference. Data scientists have expertise in computational and visualization approaches for spatiotemporal data of high dimensionality. The intersection of big data and epidemiology can change the game in terms of how we respond to pandemics like COVID-19. In this commentary, we address seven aspects of big data epidemiological studies related to the “garbage in, garbage out” of data. Next, we recommend a population level of thinking combined with spatiotemporal data analysis that has great potential to transform the fields of big data and epidemiology, respectively.