Abstract
Objectives. To evaluate public health surveillance and monitoring systems’ (PHSMS) efforts to collect, monitor, track, and analyze racism.
Methods. We employed an environmental scan approach. We defined key questions and data to be collected, conducted a literature review, and synthesized the results by using a qualitative description approach.
Results. We identified 125 PHSMS; only 3—the Behavioral Risk Factor Surveillance System, Pregnancy Risk Assessment and Monitoring System, and California Health Interview Survey—collected and reported data on individual-level racism. Structural racism was not collected in PHSMS; however, we observed evidence for linkages to census and administrative data sets or social media sources to assess structural racism.
Conclusions. There is a paucity of PHSMS that measure individual-level racism, and few systems are linked to structural racism measures.
Public Health Implications. Adopting a standard practice of racism surveillance can advance equity-centered public health praxis, inform policy, and foster greater accountability among public health practitioners, researchers, and decision-makers. Failure to explicitly address racism and the insufficient capacity to support a robust health equity data infrastructure severely impedes efforts to address and dismantle systemic racism. (Am J Public Health. 2023;113(S1):S80–S84. https://doi.org/10.2105/AJPH.2022.307160)
Public health surveillance and monitoring systems (PHSMS) collect data to guide disease prevention, improve population health, and eliminate racial/ethnic health inequities.1 Addressing inequities may be hampered, in part, by inadequate efforts to incorporate measures of racism data in PHSMS. One recent study reported the inadequacy of PHSMS in monitoring racism, stigma, and COVID-19‒related surveillance.2 However, a broader assessment of PHSMS’s capacity to collect, monitor, track, and analyze racism (operating at multiple levels) relative to general population health outcomes has not been conducted. We aimed to fill this gap by conducting an environmental scan of PHSMS to assess data collected on and linked to racism measures, highlight barriers and opportunities for data collection and linkages, and discuss public health implications.
METHODS
We performed an environmental scan to assess PHSMS capacity to collect racism data and linkages with structural racism measures. Our process entailed (1) defining key questions and data to be collected, (2) conducting a literature review, and (3) synthesizing the results with a qualitative description approach.3 A priori study questions asked (1) what are the strengths, weaknesses, and gaps in PHSMS’s capacity to collect racism and (2) to what extent are PHSMS linked with measures of structural racism? We identified PHSMS that (1) were Centers for Disease Control and Prevention–supported or –led and active from 2015 to 2020, (2) collected and reported data periodically or on an ongoing basis, and (3) monitored human health. We searched the Web site, technical documentation, data collection instruments, and publications for measures on racism, racism-related experiences, and racial discrimination. A literature review identified studies linking structural racism with PHSMS. We searched PubMed, Google Scholar, and Web of Science databases. We used a qualitative descriptive approach to synthesize the results.
RESULTS
We identified 125 PHSMS, and only 3—the Behavioral Risk Factor Surveillance System (BRFSS), Pregnancy Risk Assessment and Monitoring System (PRAMS), and California Health Interview Survey—collected and reported data on racism or race-related experiences (Table A, available as a supplement to the online version of this article at https://ajph.org).
We observed heterogeneity in measures used to operationalize racism. BRFSS collects information about health risk behaviors, conditions, and use of preventive services. Reactions to Race is an optional BRFSS module, comprising 6 questions assessing socially assigned race, race consciousness, differential treatment at work and in health care, and reports of emotional or physical symptoms to differential treatment. Since its initial pilot in 2002, approximately 50% of states administered the module for at least 1 year, with fewer states administering it in consecutive years.4 PRAMS collects data about maternal attitudes and experiences before, during, and after pregnancy. Although not a part of the “core” (fixed questions asked each year), race-related experiences 1 year before birth and during pregnancy were queried. Only 22 states assessed racism in PRAMS.5 The California Health Interview Survey provides population estimates for Californians across several health indicators. Respondents were asked about racial/ethnic discrimination in health care in select waves (i.e., 2003, 2005, 2015, 2017, and 2021).
Structural racism measures were not collected in PHSMS; however, we observed evidence for linkages to census and administrative data sets or social media sources to assess structural racism (Table 1). Data linkages enabled characterization of structural racism across judicial, economic, educational, housing, residential segregation, political, and immigration domains. Multiple quantitative measures were operationalized across each domain. For example, the economic domain included indicators related to Black‒White inequalities in unemployment, poverty, and homeownership. PHSMS most commonly linked with structural racism were BRFSS; PRAMS; National Health Interview Survey; Surveillance, Epidemiology, and End Results Program; National Death Index; and National Vital Statistics System for births, fetal deaths, and mortality data. Studies captured structural racism at multiple geographic levels including census block, census tract, zip code, county, metropolitan statistical area, and state.
TABLE 1—
Public Health Surveillance and Monitoring Systems (PHSMS) Linked to Measures of Structural Racism: United States
| Structural-Level Racism Domain | PHSMS | Operationalization | Geographic Level | Source for Data Linkage |
| Composite measure | BRFSS | Weighted estimate across criminal justice, education, employment, health care, and housing | County | US Census Bureau |
| Criminal justice | BRFSS | Police killings of unarmed Black Americans | State | Mapping Police Violence |
| Blacks’ disproportionate level of disenfranchisement | State | The Sentencing Project | ||
| Racial inequality of incarceration | State | Vera Institute of Justice | ||
| Racial inequality in juvenile custody rates | State | The Sentencing Project | ||
| Racial inequality in sentencing rates | State | The Sentencing Project | ||
| NVSS | Fatal police violence | State | Fatal Encounters Mapping Police Violence The Counted | |
| Economic | BRFSS | Racial inequality in unemployment | State County | IPUMS CPS American Community Survey |
| Racial inequality in poverty | State | IPUMS CPS | ||
| Racial inequality in median income | County | American Community Survey | ||
| Racial inequality of percentage living below the poverty line | County | American Community Survey | ||
| Education | BRFSS, SEER | Racial inequality of proportion with a bachelor’s degree | State County | IPUMS CPS American Community Survey |
| Housing | BRFSS | Racial inequality of proportion who are homeowners | State County | IPUMS CPS American Community Survey |
| SEER | Anti-Black bias in mortgage lending | Census tract | Home Mortgage Disclosure Act Data | |
| Redlining index | Metropolitan statistical area | Home Mortgage Disclosure Act Data | ||
| Immigration and border enforcement | NDI | Average of individual antiimmigrant prejudice | Metropolitan statistical area | General Social Survey |
| Political | BRFSS | Racial inequality of proportion who voted | State | US Census Bureau |
| Level of Blacks’ political underrepresentation in state legislatures | State | National Conference on State Legislatures | ||
| Residential segregation | BRFSS | Dissimilarity index | State | National Strategic Planning and Analysis Research Center |
| NHIS | Racial/ethnic residential segregation (Black‒White; Hispanic‒White) | Metropolitan statistical area | US Census Bureau | |
| SEER | Index of concentration at the extremes | Census tract | US Census Bureau | |
| Index of dissimilarity | Census block | US Census Bureau |
Note. BRFSS = Behavioral Risk Factor Surveillance System; CPS = Current Population Survey; IPUMS = Integrated Public Use Microdata Series; NDI = National Death Index; NHIS = National Health Interview Survey; NVSS = National Vital Statistics System; SEER = Surveillance, Epidemiology, and End Results Program.
DISCUSSION
Racism measures are not routinely collected and integrated in PHSMS. We identified budgetary constraints, methodological issues, decision-making authority, data linkage, and aggregation as key considerations for this observation. While a comprehensive racism measure may assess chronicity, recurrence, severity, and duration, and delineate between direct and indirect experiences,6 concerns about survey length may constrain the type of scales included. The decision-making authority that determines and gives value to the data included in PHSMS raises serious equity issues. For example, state BRFSS advisory committees composed of community and academic partners who provide input and may bear financial responsibility for items administered in optional modules. This can lead to bias and the continued omission of racism measures in PHSMS. Actionable suggestions to address these data gaps entail adding racism measures to a rotating core. The permanent adoption of racism measures as standard fixed questions would mirror the recent decision by PRAMS leadership and set a poignant standard for PHSMS.5,7
In synthesizing findings from PHSMS linked with structural racism, studies leveraged multiple data sources, social media, innovative tools for data generation, and data-mining techniques (e.g., machine learning) to operationalize structural racism. Other novel opportunities to characterize structural racism and link with PHSMS involve designing data clearinghouses for historical and contemporary laws8 and research tools that permit access to data sharing across government agencies. For example, the New Jersey Integrated Population Health Data Project develops an integrated data system linking health and social administrative data.9
PUBLIC HEALTH IMPLICATIONS
Integrating racism as a sentinel indicator in PHSMS can advance equity-centered public health praxis and antiracist policy development, implementation, and evaluation, and foster greater accountability among public health actors. The absence of racism data precludes the development of data-driven health objectives and hampers targeted evidence-based action to address health inequities. For example, the Healthy People initiative guides national health promotion, disease prevention, and health equity efforts. Every 10 years, data from PHSMS are used to inform measurable objectives and set benchmarks to evaluate progress. While select social determinants of health (e.g., educational attainment) are tracked and considered targets for action, capturing the lived experience of racism with the same scientific rigor and consistency is nonexistent.
Harnessing racism data has the potential to strengthen data-driven governance and data-based policymaking to create equitable communities. Racism data coupled with racial equity tools (e.g., racial equity impact assessments) can be used to critically evaluate the effect of budgetary decisions, policies, legislation, and regulations on population health and inequities. Systems of accountability for public health practitioners, health care providers, policymakers, and other key stakeholders can be designed. For example, a novel structural racism measure utilizes data from the Census Bureau’s Census of Governments, which collects information on financial decision-making related to revenues, expenditures, debts, and assets across government entities.10 These data can illuminate structural forces influencing financial decision-making.
Antiracist public health necessitates an infrastructure with data tools that collect, track, and evaluate dynamic patterns of racism at all levels. Advancing health equity requires strategies for sustained political support and systemic change. For example, in 1992, Congress passed the Cancer Registries Amendment Act (Pub L No. 102–515) establishing the National Program of Cancer registries, which authorized funds to develop and set standards for cancer registries and establish a reporting system.11 Earmarking funds to finance the optimization of PHSMS to capture, analyze, and report racism data would represent an intentional effort toward equity-centered surveillance. There is a collective memory in communities that endures the scars of the unethical use of data that reifies racist ideologies and perpetuates intergenerational racial/ethnic health inequalities. While data alone will not serve as a panacea for dismantling racism, the omission to explicitly name, measure, collect, and track racism data severely impedes science and precludes translational efforts to achieve health equity.
ACKNOWLEDGMENTS
The authors acknowledge Blen Asres and Alfonso Rodriguez-Lainz for help with identifying the public health surveillance and monitoring systems and David R. Williams and Camara Phyllis-Jones for helpful comments on a preliminary draft of the article.
CONFLICTS OF INTEREST
The authors declare no conflicts of interest.
HUMAN PARTICIPANT PROTECTION
Institutional review board approval was not required for this study because it does not meet the criteria of human participant research.
REFERENCES
- 1.Groseclose SL, Buckeridge DL. Public health surveillance systems: recent advances in their use and evaluation. Annu Rev Public Health. 2017;38(1):57–79. doi: 10.1146/annurev-publhealth-031816-044348. [DOI] [PubMed] [Google Scholar]
- 2.Ford CL, Amani B, Harawa NT, et al. Adequacy of existing surveillance systems to monitor racism, social stigma and COVID inequities: a detailed assessment and recommendations. Int J Environ Res Public Health. 2021;18(24):13099. doi: 10.3390/ijerph182413099. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Charlton P, Doucet S, Azar R, et al. The use of the environmental scan in health services delivery research: a scoping review protocol. BMJ Open. 2019;9(9):e029805. doi: 10.1136/bmjopen-2019-029805. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Srivastav A, Robinson-Ector K, Kipp C, Strompolis M, White K. Who declines to respond to the reactions to race module?: Findings from the South Carolina Behavioral Risk Factor Surveillance System, 2016–2017. BMC Public Health. 2021;21(1):1703. doi: 10.1186/s12889-021-11748-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Almeida J, Belanoff C, Erbetta KF. The time has come for all states to measure racial discrimination: a call to action for the Pregnancy Risk Assessment Monitoring System (PRAMS) Matern Child Health J. 2022;26(1):7–11. doi: 10.1007/s10995-021-03160-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Williams DR. Improving the measurement of self-reported racial discrimination: challenges and opportunities. In: Alvarez AN, Liang CTH, Neville HA, editors. The Cost of Racism for People of Color: Contextualizing Experiences of Discrimination. Washington, DC: American Psychological Association; 2016. pp. 55–83. [DOI] [Google Scholar]
- 7.D’Angelo DV, Dieke A, Williams L, et al. Response to “The time has come for all states to measure racial discrimination: a call to action for the Pregnancy Risk Assessment Monitoring System (PRAMS. Matern Child Health J. 2022;26(1):12–14. doi: 10.1007/s10995-021-03279-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Agénor M, Perkins C, Stamoulis C, et al. Developing a database of structural racism-related state laws for health equity research and practice in the United States. Public Health Rep. 2021;136(4):428–440. doi: 10.1177/0033354920984168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Rutgers Center for State Health Policy. 2022. https://iphd.rutgers.edu/about
- 10.Hardeman RR, Homan PA, Chantarat T, Davis BA, Improving THB. The measurement of structural racism to achieve antiracist health policy. Health Aff (Millwood). 2022;41(2):179–186. doi: 10.1377/hlthaff.2021.01489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Izquierdo JN, Schoenbach VJ. The potential and limitations of data from population-based state cancer registries. Am J Public Health. 2000;90(5):695–698. doi: 10.2105/AJPH.90.5.695. [DOI] [PMC free article] [PubMed] [Google Scholar]