The Role of Health Equity in Quantitative Research

Disparities in key areas such as access to education, housing, healthcare and employment have been pervasive in many societies. Adverse events, including natural disasters and health emergencies, disproportionately impact populations who are at increased risk, further exacerbating disparities. For example, the COVID-19 pandemic illustrated how communities facing barriers have endured magnified challenges and hardships.

There is a growing use of data along with operations research (O.R.) and analytics tools to address societal challenges and policy decisions in healthcare. Such quantitative approaches must carefully consider disparities and equity not only in the allocation of limited resources but also in the resulting health outcomes.

Health disparities are defined as “a particular type of health difference that is linked with social, economic, and/or environmental disadvantage [that] adversely affect groups of people who have systematically experienced greater obstacles to health …” [1]. Factors such as race and ethnicity, gender, geography, sexual orientation, education and income level impact health outcomes. For example: (1) The maternal death rate for non-Hispanic (NH) Black women is 3.55 times higher than for NH white women [2]. (2) The prevalence of lung cancer is higher in rural versus urban populations in the U.S. [3]. (3) Hispanic and Black children have higher risk of asthma incidence and earlier onset compared with white children [4]. Although biological factors might play a role, many of these disparities are in part due to upstream factors; for example, racial and ethnic minority groups and populations with lower income are more likely to live in areas with unhealthy levels of particulate matter pollution [5], which increases their chances of developing asthma.

Health equity acknowledges that individuals and communities have different vulnerabilities and health risks, socioeconomic factors, healthcare needs and access to key resources and services. The prevalence and extent of health disparities prompted efforts to achieve health equity by addressing the upstream factors, recognizing that equality (in access, resource allocation, etc.) is not always equitable. Lived experiences and other factors impact peoples’ abilities and attitudes in engaging with healthy behaviors and healthcare systems. Therefore, interventions and solutions need to consider heterogeneity across subpopulations and caution against a one-size-fits-all approach.

Integrating Health Equity into Data-driven Research and Decision-making

As O.R. and analytics professionals, we develop and creatively apply quantitative methods to support policy and managerial decisions. Health systems are complex, with numerous stakeholders and their corresponding (sometimes conflicting) incentives. To address health disparities, it is important to incorporate systems thinking into every phase and aspect of O.R./analytics activities: from data collection and analysis to solution approaches and evaluation metrics. 

“Missingness” in health data is typically neither random nor uniform across subpopulations; it often correlates with health disparities and, if neglected, can further increase health inequities. For example, during our collaborations with the Georgia Department of Public Health in developing COVID-19 health equity-related dashboards [6], we noticed that the state and/or county of residence information was missing in 17.7% of COVID-19 vaccination records for Hispanic adults versus 3.7%-7.4% in other races and ethnicities. When all records without residence information were excluded, Hispanic adults had the lowest primary series vaccination rate (52.6%) compared with NH Asian (75.9%), NH white (55%) and NH Black adults (53.6%), from December 2020 to February 2023 in Georgia. However, when the records with missing residence data were included, Hispanic adults had the second largest vaccination rate at 64%, only behind NH Asian adults (82%) [7]. Hence, careful analysis of how and why data missingness varies across subpopulations can inform equitability in future data collection efforts as well as reduce bias in the findings and recommendations resulting from the analysis.

Central to addressing health equity is the identification of metrics and objectives that holistically capture the multidimensional nature of disparities while concurrently ensuring solution effectiveness. For example, consider the “facility location” problem within the public health context – e.g., selecting the locations for vaccination sites during an epidemic or pandemic to provide equitable access to these sites. Many factors impact access, such as physical proximity, affordability, availability, quality and ability to serve individuals with different backgrounds and needs. Considering only one (or a few) of these metrics might not lead to equitable access. For example, maximizing the percentage of the population with access to at least one vaccination site within a 10-minute drive might seem equitable at first but could disadvantage individuals who rely on public transport.

Even in cases in which we achieve equitability in the allocation of or access to a particular set of resources, this does not always lead to equitability in outcomes, unless we achieve equitability across multiple areas, such as access to preventive and curative care, employment benefits, etc. For example, during a pandemic, distributing and administering vaccines proportional to the population might not lead to equitable health outcomes. Our analysis of COVID-19 vaccination and outcomes in Georgia showed that even when controlling for vaccination, NH Black adults had higher death and hospitalization rates than their counterparts in other racial groups [7]. Hence, in resource allocation and prioritization, it is important to holistically consider the attributes of different subpopulations to improve equitability in outcomes.

Decision-making in healthcare is complex, given the impact of many upstream factors affecting equity in health outcomes and the need to balance multiple objectives. The decisions need to be updated dynamically because they impact human behavior and system dynamics, which in turn change subsequent decisions in resource allocation, interventions and strategies. Crucial to the pursuit of timely, responsive and effective data-driven decision-making for health equity is joining forces and establishing synergistic collaborations with medical and public health professionals and social scientists, among others. There is mounting evidence and much promise in further development of new O.R. and analytics methods and their innovative applications in public health with an equity lens, to address the complex and changing needs of diverse communities, leading to happier and healthier lives.

References

1. S. Department of Health and Human Services, 2022, “Health Equity and Health Disparities Environmental Scan,” Rockville, MD: U.S. Department of Health and Human Services, Office of the Assistant Secretary for Health, Office of Disease Prevention and Health Promotion, https://health.gov/sites/default/files/2022-04/HP2030-HealthEquityEnvironmentalScan.pdf.
2. MacDorman, M. F., Thoma, M., Declcerq, E. & Howell, E. A., 2021, “Racial and ethnic disparities in maternal mortality in the United States using enhanced vital records, 2016‒2017,” American Journal of Public Health, Vol. 111, No. 9, pp. 1673-1681.
3. NIH National Cancer Institute, “Cancer map stories,” GIS Portal for Cancer Research - Rural Urban Disparities in Cancer, https://gis.cancer.gov/mapstory/rural-urban/index.html.
4. Zanobetti, A., Ryan, P. H., Coull, B., Brokamp, C., Datta, S., Blossom, J., et al., 2022, “Childhood asthma incidence, early and persistent wheeze, and neighborhood socioeconomic factors in the ECHO/CREW consortium,” JAMA Pediatrics, Vol. 176, No. 8, pp. 759-767.
5. Jbaily, A., Zhou, X., Liu, J., Lee, T. H., Kamareddine, L., Verguet, S. & Dominici, F., 2022, “Air pollution exposure disparities across US population and income groups,” Nature, Vol. 601, No. 7892, pp. 228-233.
6. Fujimoto, A., Keskinocak, P. & Nazzal, D., 2023, “COVID-19 Dashboards and Decision-Support Tools,” Center for Health and Humanitarian System, https://chhs.gatech.edu/covid19-dashboard.
7. Fujimoto, A., Edison, L., Keskinocak, P. & Nazzal, D., 2023, “Racial/ethnic disparities in adult COVID-19 vaccination, deaths, and hospitalizations in Georgia," Forthcoming.