Understanding the impact of COVID-19 pandemic on social determinants of health of racial and ethnic minorities using natural language processing and machine learning

ABSTRACT

The COVID-19 pandemic has contributed to a sharp increase in health disparities among racial and ethnic US minority communities. This study aims to understand the social determinants of health issues of racial/ethnic US minority populations before and during the COVID-19 pandemic. Specifically, this study establishes the extent to which Latent Dirichlet Allocation (LDA) and Gibbs Sampling Dirichlet Multinomial Mixture (GSDMM)-based topic modeling determines social determinants of health (SDOH) categories, and how adequately custom named-entity recognition (NER) detects key SDOH factors from a Reddit corpus. Moreover, we performed an inductive thematic discourse analysis on the Reddit corpus and High Density-Based Spatial Clustering of Applications with Noise (HDBSCAN)-based clustering on the thematic discourse results. We collected race/ethnicity-related data from five subreddits representing five highly populated cities in the US from March to December 2019 (prior to the pandemic) and from March to December 2020 (during the pandemic). Our study identified 35 SDOH-related topics, 22 themes, and revealed that the COVID-19 pandemic significantly exacerbated SDOH issues of racial/ethnic US minority communities. On average, conversations about the Social and Community Context (SCC) category of SDOH had the highest percent increase (358%) from the pre-pandemic period to the pandemic period across all locations and population groups. Some of the SCC issues were racism, protests, arrests, immigration, police brutality, hate crime, white supremacy, and discrimination. The dissertation offers automatic ways to glean SDOH-related information from social media. Furthermore, the knowledge gained from this study will empower researchers, governments, and policymakers to design interventions to shift racial and ethnic disparities toward more equitable outcomes. (PsycInfo Database Record (c) 2023 APA, all rights reserved)