Leveraging Online Big Data to Assess the Impact of Macro-Level Factors on Health

Abstract

Health inequalities are well documented to parallel social inequities. Although a large body of research has investigated the interpersonal and individual level social factors that influence health outcomes, there has been less emphasis on the macro-level and structural factors that foster and contribute to social inequities and disparities in health. With the rise in use of digital technology, social media, and web platforms, there is now a wealth of digital and online data that can potentially capture the impact of macro-level factors on health that were previously not possible to monitor using existing data sources. This dissertation combines traditional epidemiology and novel computer science methods to examine the effects of the broad and societal factors that impact social inequities in health. In study 1, a longitudinal 3-level multilevel hierarchical analysis across time, county and state is conducted to investigate the impact of Jim Crow laws on social capital from 1997-2014. In study 2, online big data from social media is collected to monitor racial disparities in patient experience in the United States from 2013-2016. In study 3, online big data is leveraged to conduct a lag-time series to identify the impact of racial animus on mental health outcomes of vulnerable populations during the 45th presidential election. Findings from these studies showcase the feasibility of using online big data to uncover macro-level factors that influence disparities in health. Historical, societal and structural factors have shaped the distribution of social inequities and the persistence of inequitable health outcomes. These studies provide evidence of the relationship between macro-level factors and disparities in population health.