Examining Racial Discrimination Index and Black-Years of Potential Life Lost (YPLL) in South Carolina: A Real-Time Social Media Research

Journal of Racial and Ethnic Health Disparities https://doi.org/10.1007/s40615-025-02416-7 Examining Racial Discrimination Index and Black‑Years of Potential Life Lost (YPLL) in South Carolina: A Real‑Time Social Media Research Yunqing Ma1 · Peiyin Hung1 · Xiaotong Shen2 · Zhenlong Li3 · Curisa Tucker4 · Jiajia Zhang1 Received: 5 November 2024 / Revised: 26 March 2025 / Accepted: 29 March 2025 © The Author(s) 2025 Abstract Purpose Despite efforts to reduce health disparities, Black Americans still face higher mortality rates than Whites. Racism has been recognized as a significant social determinant of health. Using social media data, human-being qualitative coding, and AI, we created a county-level racial discrimination index, exploring its association with years of potential life lost (YPLL) rates. Methods Through human-AI collaborative approaches using X/Twitter data, we calculated yearly county-level racial discrimination index (RDI)—number of racial discrimination posts per 100,000 in-county non-duplicated posts and examined the relationship between RDI terciles and YPLL per 100,000 non-Hispanic Black individuals. Annual data for the covariates were derived from 2018–2022 County Health Rankings and American Community Surveys across all South Carolina (SC) counties. Results RDI increased from 2018 (mean [SD], 1.443 [1.991]) to 2022 (3.439 [5.761]). Adjusting for county sociodemographic factors and historical trends, RDI was associated with the YPLL rate (marginal effects, highest vs. lowest tercile, 421.3; 95% confidence interval, 134.7–709.8; p = 0.006). Conclusions Digital racial discrimination was highly associated with Black YPLL rates, confirming the importance of racial discrimination in health disparity, especially premature deaths. Addressing explicit and implicit racism in highly affected counties is crucial for reducing persistent health inequities and promoting equity in communities. Keyword Racial discrimination index · YPLL · South Carolina · Social media Introduction In 2000, the Institute of Medicine (IOM) published the landmark report “To Err is Human,” one of the earliest acknowledgments, which brought national attention to systemic issues in the US healthcare system, including the role of racism in shaping racial health disparities [1, 2]. More * Peiyin Hung 1 Arnold School of Public Health, University of South Carolina, 915 Greene Street, Columbia, SC 29208, USA 2 School of Statistics, University of Minnesota, Twin Cities, USA 3 Department of Geography, The Pennsylvania State University, State College, PA 16801, USA 4 College of Nursing, University of South Carolina, Columbia, SC 29208, USA than two decades later, Black Americans still experience disproportionately worse health outcomes, such as premature death [3, 4]. Over the past 20 years, Black Americans have experienced 1.63 million excess deaths and lost over 80 million years of life in contrast to their non-Hispanic White counterparts [3, 5]. Particularly, the average rate of years of potential life lost (YPLL) per 100,000 individuals over the past 22 years was 20,365 for White males and 15,428 for White females, while for Black males and females, it stood at 31,944 and 23,360, respectively [3, 6]. Persistent premature death disparities between Black and White Americans are multifaceted and often result from structural racism and discrimination [7]. Rooted in slavery, segregation, and ongoing racial biases, structural racism continues to shape health disparities facing Black populations [8–10], by limiting their access to resources and opportunities. Often, Black individuals were less likely than their White peers to have access to quality education for adequate health literacy, stable employment, safe housing, Vol.:(0123456789) Journal of Racial and Ethnic Health Disparities healthy food, and timely and culturally competent healthcare [11]. The scarcity of these resources and opportunities that promote health, combined with discrimination entrenched in some communities within American society, further perpetuated avoidable death and widen Black-White disparities in health outcomes and YPLL [12, 13]. Racism, in its essence, refers to the belief in the inherent superiority or inferiority of individuals based on their race, coupled with actions or policies that reinforce these beliefs [14]. It manifests in various forms, ranging from interpersonal racism, institutional racism, to structural racism [15–17]. Existing measures of racism focus on structural racism, such as Residential Segregation Index, and Structural Racism Effect Index. Specifically, Residential Segregation, which reflects the likelihood of Black residents interacting with White residents across Census tracts in a county, is commonly used to serve as a proxy for structural racism due to historical oppression [18]. The Residential Segregation Index quantifies this segregation by providing a measurable indicator of racial separation within communities. Structural Racism Effect Index (SREI) was established to measure the neighborhood-level impact of structural racism across the United States using publicly available data, including the American Community Survey (ACS), the U.S. Census Bureau Supplemental Poverty Measure, and the National Center for Health Statistics U.S. Small-area Life Expectancy Estimates Project (USALEEP) [19]. However, the conceptualization and operationalization of racism in health research often exhibit gaps, primarily due to the reliance on self-reported perceived discrimination or historical structural factors [20, 21]. Self-reported measures of perceived discrimination, while valuable in capturing individuals’ experiences, are inherently subjective and susceptible to various biases, including cognitive biases and social desirability [22]. Moreover, such measures typically only capture overt acts of discrimination that are consciously recognized by individuals, potentially overlooking more subtle forms of bias and systemic inequalities [22]. Additionally, self-reported measures of discrimination may lack the contextual information necessary to fully comprehend the underlying social and economic factors contributing to racial disparities in health [22, 23]. As a result, there is a need for more nuanced and comprehensive approaches to conceptualizing and measuring racism in health research, one that considers both individual experiences and broader structural determinants of health outcomes. With social media being widely used, it has become a leading approach for communication and information exchange, despite ongoing debates regarding its authenticity and psychological impact [24, 25]. While social media platforms can sometimes present curated or distorted portrayals of life, they also serve as spaces where users share candid thoughts, engage in public discourse, and report personal experiences in real time, making them valuable data sources for social research. The utilization of social (...truncated)