Skip to main content
In their article, Lett et al eloquently discuss the limitations of using race as a proxy for structural and individual racism and make recommendations for the public health and medical communities to consider when designing health sciences research (1). One of their recommendations includes using information from outside data sources (examples given include the US Census Bureau and Centers for Disease Control and Prevention (CDC) databases) to better understand the differences seen between racial and ethnic groups within the context of other measures of structural racism. The goal of this reply is to explore how disadvantage indices (DI) may help further this goal.
DIs are measures designed to capture the relative advantage or disadvantage of groups of people based on a subset of metrics within a defined geography. Two of the most commonly used examples of DIs include the CDC’s Social Vulnerability Index (SVI) (2) and the Area Deprivation Index (ADI) (3,4). The SVI and ADI both characterize four domains (or themes) of social vulnerability based on combinations of census variables. The SVI measures socioeconomic status, housing type/transportation, household characteristics, and racial and ethnic minority status per unit of geography. While similar, the ADI uses census variables to characterize the themes of education, income/employment, housing, and household characteristics. Many of the included variables, if not all, are in some way influenced by the forces of individual and structural racism (5,6).
DIs were originally created as a tool for disaster management to identify areas that would need additional support in the setting of a man-made or natural disaster (2), but are well positioned to explore social vulnerability and public health outcomes. DIs can be used on the population-level to compare health outcomes by geographic region or individual patient-level data can be mapped to DI scores. Studies using DIs have demonstrated worsening health outcomes for racial minority groups living in high socially disadvantaged areas, illustrating the utility of utilizing more robust constructs of social vulnerability to capture the impact of structural and individual racism in differential health outcomes (7-9).
Increasing the measurement of indicators of individual and structural racism improves the ability quantify the impact of racism. Compiling such indicators into a score has the potential to account for the co-occurrence and compounding intersectionality of indicators rather than controlling for each individual metric when measured individually. It is this compilation of a variety of variables into a single number for analysis that makes DIs an improved tool for incorporating a nuanced measure of social vulnerability into health sciences research. That said, DIs can also be leveraged based on data availability, with researchers having the option to use subsets of DI metrics that may better fit the testing of a specific conceptual model, or if appropriate, exploring individual variables of interest.
Some limitations of DIs should be considered. First, DIs use variably sized geographic areas which increases issues of heterogeneity within the sample (4). For example, while the ADI uses census block group (average size 600 to 3,000 people) to represent geography, the SVI’s smallest geographic area is the census tract (average size around 4,000 people) (10), resulting in a lower geographic specificity to associations. Additionally, DIs have considerable heterogeneity between their included variables (4). Researchers need to have an understanding of the components of the DI that they are using in order to comment on ways the choices of variables making up the DI map to conceptual models of individual and structural racism as well as the impact on bias.
There have been efforts both within the most common DIs and through development of new DIs to address these limitations. Notably for the SVI, the most recent dataset (2020) recently updated the subtheme “Racial & Ethnic Minority Status.” Specifically, they moved the “English Language Proficiency” variable out of the subtheme and into “Household Characteristics” subtheme due to the concern that it was “adversely impacting the vulnerability ranking in communities of high minority areas of the country,” recognizing that while “people in racial and ethnic minority groups are overall more likely to have limited English language proficiency than non-Hispanic whites, most (90.9%) are English language proficient.” They also updated the theme to include individual race categories, which was not previously included (11). While these are small changes, they demonstrate a commitment to continually improve the utility of the dataset to capture health inequity. Additionally, entirely new indices have been compiled, such as the Minority Health SVI. The Minority Health SVI combines the variables in the original SVI with additional census variables, including healthcare specific measures such as hospitals/urgent care/pharmacies per 100,000 and chronic disease rates (diabetes, respiratory disease, cardiovascular disease), as well as expanded race/ethnicity and language variables (12).
As outlined, DIs may be one tool to move away from solely using race as a proxy measure for racial disparity and towards capturing the complex ways that individual and structural racism create health inequity. In particular, when used in conjunction with race and ethnicity data, DIs can help demonstrate the impact of individual and structural racism in comparing data between groups situated in similar geographic regions. DIs have important limitations and, as recommended in Lett et al, efforts to improve them should be undergone under the guidance and leadership of Black, Latine, and Indigenous scholars. Additionally, they should not be seen as a solution by themselves, but as a tool to use alongside the other recommendations made by Lett et al.
REFERENCES
1. Lett E, Asabor E, Beltran S, Cannon AM, Arah OA. Conceptualizing, Contextualizing, and Operationalizing Race in Quantitative Health Sciences Research. Ann Fam Med. 2022;20(2):157-163.
2. Barry E Flanagan EWG, Elaine J Hallisey, Janet L Heitgerd, Brian Lewis. A Social Vulnerability Index for Disaster Management. Journal of Homeland Security and Emergency Management. 2011;8(1).
3. Kind AJH, Buckingham WR. Making Neighborhood-Disadvantage Metrics Accessible - The Neighborhood Atlas. N Engl J Med. 2018;378(26):2456-2458.
4. Srivastava T, Schmidt H, Sadecki E, Kornides ML. Disadvantage Indices Deployed to Promote Equitable Allocation of COVID-19 Vaccines in the US. JAMA Health Forum. 2022;3(1):e214501.
5. Khazanchi R, Evans CT, Marcelin JR. Racism, Not Race, Drives Inequity Across the COVID-19 Continuum. JAMA Netw Open. 2020;3(9):e2019933.
6. Yearby R, Mohapatra S. Law, structural racism, and the COVID-19 pandemic. J Law Biosci. 2020;7(1):lsaa036.
7. An R, Xiang X. Social Vulnerability and Leisure-time Physical Inactivity among US Adults. Am J Health Behav. 2015;39(6):751-760.
8. Killian AC, Shelton B, MacLennan P, et al. Evaluation of Community-Level Vulnerability and Racial Disparities in Living Donor Kidney Transplant. JAMA Surg. 2021;156(12):1120-1129.
9. Dalmacy DM, Tsilimigras DI, Hyer JM, Paro A, Diaz A, Pawlik TM. Social vulnerability and fragmentation of postoperative surgical care among patients undergoing hepatopancreatic surgery. Surgery. 2022;171(4):1043-1050.
10. Our Surveys & Programs: Geography Program: About Us: Glossary. United States Census Bureau. https://www.census.gov/programs-surveys/geography/about/glossary.html#:~...(BGs)%20are%20statistical,data%20and%20control%20block%20numbering. Published 2022. Updated April 11, 2022. Accessed January 1, 2023, 2023.
11. CDC/ATSDR SVI 2020 Documentation. https://www.atsdr.cdc.gov/placeandhealth/svi/documentation/pdf/SVI2020Do.... Published 2020. Accessed January 1, 2023, 2023.
12. Minority Health SVI. U.S. Department of Health and Human Services, Office of Minority Health. https://www.minorityhealth.hhs.gov/minority-health-svi/. Accessed January 2, 2023, 2023.