Social Determinants of Health and Estimation of Kidney Function

In 1999, the Modification of Diet in Renal Disease (MDRD) Study Group published the original MDRD Study equation to estimate GFR on the basis of serum creatinine concentration. The equation included a coefficient for Black race because Black race was associated with higher serum creatinine levels and less precise estimates of GFR.¹ More recently, investigators from the Chronic Renal Insufficiency Cohort (CRIC) study reported that the use of serum creatinine to estimate GFR without adjusting for Black race (or African ancestry) led to systematically biased estimates of GFR among Black participants.² By contrast, studies in diverse populations outside the United States (e.g., Africa and South America) have reported that adjusting for Black race actually showed greater bias and lower accuracy of GFR estimates when using serum creatinine–based equations.³ The inconsistent associations of Black race and serum creatinine–based GFR estimates could be attributable to differences in nonglomerular factors that influence creatinine (e.g., muscle mass, dietary protein intake, tubular secretion, and nonrenal elimination). However, the observed differences could also be attributable to a much wider array of forces and systems that correlate with Black race and racism in America, which also shape nutrition, physical activity, and health through conditions of daily life—social determinants of health.⁴

In the United States, discriminatory policies have created marked inequities in the social environment, as evidenced by racially and economically segregated communities; differential access to high-quality schools, nutritious food, and safe housing; and marked variation in neighborhood walkability, availability of green spaces, and exposure to environmental pollutants.⁴ In this issue of CJASN, Eneanya et al.⁵ apply a social determinants framework to investigate the influence of selected sociodemographic factors—individual-level educational attainment, employment status, marital status, and annual household income—on the performance of serum creatinine–based GFR‐estimating equations. The investigators re-examined data from the MDRD and CRIC studies to test whether measured sociodemographic factors and nonglomerular determinants of creatinine could account for differences in serum creatinine–based GFR estimates previously attributed to Black race.

The investigators observed significant between-study variation in the magnitude of difference in measured GFR versus serum creatinine–based estimated GFR (21% in MDRD versus 13% in CRIC) attributed to Black race. The investigators also found that sociodemographic factors (educational attainment, employment status, marital status, and household income) differed between Black and non-Black participants in both studies and between Black participants from MDRD and Black participants from CRIC. However, adjustment for education, employment status, marital status, and household income did not materially influence the statistical relationship of Black race and serum creatinine–based GFR estimates in either study. The between-study differences observed by Eneanya et al. could plausibly reflect inconsistencies in the categorization of race. However, these differences in the coefficient for Black race more likely represent variation in human biology and social experiences that the study variables ascertained in the MDRD and CRIC studies did not adequately capture.

Nephrology has a long history of linking socially defined race to disease. However, these observed associations rarely account for cultural, social, and structural determinants of health that correlate with race, racism, and health in America as framed by Eneanya et al. Their study draws further strength from the assessment of creatinine methodologies of MDRD and CRIC, the latter of which concurrently measured GFR; serum creatinine; urinary creatinine excretion; and nonglomerular determinants of serum creatinine, such as estimates of dietary protein intake and measures of body composition.^1,2 Perhaps, the most pronounced study limitation reflects our scientific community's historical indifference to studying and addressing the unmet social needs of our patients and to examining the effect of structural racism on kidney health. To the same extent that a binary race variable cannot possibly capture human biologic diversity, the sociodemographic variables examined (individual-level education, employment status, marital status, and household income) represent but a limited cross-section of myriad social, cultural, and political upstream factors that influence a person's health and ability to live a healthy life.

In looking forward, the study by Eneanya et al. necessitates a more inclusive and systematic approach to incorporating social determinants of health into nephrology research and clinical practice. As the investigators note, a “wider array of factors…may reflect the effects of systemic racism and discrimination on health.”⁵ Racism—organized systems within our society that result in avoidable, unequal distribution of power, resources, and opportunities on the basis of race—frequently affects health through social conditions. Although the relative contributions may vary, studies overwhelmingly find that social determinants collectively and consistently contribute far more to health and health outcomes than medical care. However, a pervasive challenge of social determinants of health research remains in how to translate population-based observations into discrete, actionable interventions. To start, it is imperative that clinical studies consistently collect data that more robustly capture health-related social determinants to better support approaches such as that of Eneanya et al.

Nephrology historically adopts a mechanistic lens to isolate pathways that lead to concrete outcomes in disease-specific models. By contrast, the study of social determinants aligns more readily with a multilevel approach, where researchers and policy makers must consider the complex interplay of factors that link to health outcomes at multiple interconnected levels. The socioecological model, which frames health disparities across policy, community, organization, interpersonal, and individual levels, offers a structure to evaluate differential outcomes previously ascribed to race or ethnicity. Adopting such a framework enables a systematic approach to explore the complex effect of racism on kidney health. When examined through a socioecological lens, the work by Eneanya et al. includes only individual and interpersonal variables and thus incompletely investigates many of the social determinants of health.

Importantly, the study of social determinants of health presents a strategic opportunity for kidney researchers, clinicians, and health system leaders to collaborate with community partners across different care settings. Despite decades of studies linking social, economic, and political factors to racial disparities in the incidence and treatment of end‐stage kidney disease,⁶ our understanding of the care structure and unmet health needs of historically marginalized groups in earlier stages of kidney disease remains limited. Notably, many community health centers—the backbone of the American health care safety-net—have systematically screened for health-related social needs and developed and coordinated social services with community partners to address the identified needs of their patients for years.⁷ For example, the National Association of Community Health Centers uses the Protocol for Responding to and Assessing Patients' Assets, Risks, and Experiences tool to collect demographic information and screen for various social needs, including housing, employment, education, security, transportation, social integration, and stress.⁷ Although the use of screening tools to assess health-related social needs has increased in recent years, the validity and reliability of different screening tools, including how these tools perform in different populations, remains unclear. Research at multiple levels could accelerate by adopting a standardized approach to health-related social needs screening and automating processes to integrate results as structured data. In addition, the study of large-scale data (e.g., street-level images) and their relations to kidney health may provide additional insight into neighborhood inequalities and how structural racism and the built environment influence kidney disease outcomes.⁸

Diverse perspectives—within and outside of nephrology—serve as the foundation for attaining equitable kidney health. In 2019, medical students at the University of Washington questioned the use of race in estimating GFR.⁹ Unlike serum creatinine, for which standardized measurements exist to minimize bias between laboratories, the students highlighted that race is a sociopolitical construct that eludes measurement by physical features or genetic markers. Last year, several national kidney disease organizations recommended the adoption of race-free equations to estimate GFR¹⁰—a step forward in differentiating the erroneous conflation of biologic entity and social identity. The study by Eneanya et al. highlights additional challenges toward health equity in nephrology, starting with addressing bias in health data and continuing with improving tools and methods to capture inequities in the social and structural determinants of kidney health. To do so, the nephrology community must embrace the study of racism and social determinants of kidney health as an important area for research as variation in serum creatinine.

Disclosures

J.L. Bullock has received speaking honoraria from the American Academy of Pediatrics, Ascension Health, Baylor Scott & White Health, Christus Health, Kaiser Foundation, the Macy Foundation, Ohio State University, University of Iowa, University of Wisconsin‐Madison, and USMLE. Y.N. Hall reports employment with Veterans Affairs Puget Sound Health Care System and Kidney Research Institute, University of Washington; serving on the Editorial Board of American Journal of Kidney Diseases (unpaid); and other interests or relationships with Northwest Kidney Centers (prior Medical Director).

Funding

None.

Author Contributions

Conceptualization: Justin L. Bullock, Yoshio N. Hall.

Project administration: Yoshio N. Hall.

Supervision: Yoshio N. Hall.

Writing – original draft: Justin L. Bullock, Yoshio N. Hall.

Writing – review & editing: Justin L. Bullock, Yoshio N. Hall.