Higher socioeconomic status is not associated with equal reductions in rates of type 2 diabetes and obesity across all racial and ethnic groups in the United States, according to a new study published July 8, 2026, in the open access journal PLOS One by Sara Cromer of Harvard Medical School, USA, and colleagues.
Unfavorable socioeconomic status (SES) is associated with adverse health outcomes, and many researchers, clinicians, and risk calculators assume that improvements in SES confer equivalent health benefits across all population groups.
In the new study, the researchers analyzed data from 54,991 adult participants in the US National Health and Nutrition Examination Survey (NHANES) from 1999–2018 alongside 404,990 participants in the All of Us (AoU) cohort. They assessed the associations between educational attainment, income, and prevalence of type 2 diabetes (T2D) and obesity.
Age-adjusted rates of T2D and obesity were generally highest in participants with lower SES, and among non-Hispanic Black, Mexican American, Other Hispanic, and Other/Multi-Racial participants. When examining how rates varied with SES in different racial and ethnic groups, using adjusted models, the researchers found that higher educational attainment was associated with a 12% reduction in T2D prevalence among non-Hispanic White participants (OR 0.88, 95% CI 0.85–0.91) but only a 4% reduction among non-Hispanic Black participants (OR 0.96, 95% CI 0.92–0.99). Similarly, higher income was protective against obesity among non-Hispanic White participants (OR 0.97, 95% CI 0.95–0.996) but associated with higher rates of obesity among non-Hispanic Black participants (OR 1.05, 95% CI 1.01–1.08). Effects also varied by data set and by the SES measure used - education associates with metabolic disease differently than income, for example, and these related measures capture different risks.
The study's cross-sectional design means that it cannot prove causation, and the AoU cohort is subject to selection bias, with non-Hispanic Asian participants in particular showing very high levels of educational attainment. Despite these limitations, the authors conclude that the direction, magnitude, and shape of the association between SES and metabolic disease are heterogeneous across US racial and ethnic groups and that treating these associations as uniform may obscure important differences, particularly among minoritized populations.
Sara Cromer adds: "Although improving education or income is associated with lower rates of diabetes and obesity on average, the amount of improvement varies significantly by race. Overall, Black, Hispanic, and Asian individuals in the US have decreased protective effects of higher education and income than White individuals."
"We have known for many years that decreased education and income are associated with worse health outcomes. Understanding how education and income interact with health across all communities is increasingly important right now because, in efforts to advance precision medicine but also prevent race-based medical decision-making, medical risk calculators are starting to include measures like education and income to help doctors make treatment recommendations. If we don't have a good understanding of how these factors impact health, or if we ignore differences across communities, these calculators may lead to worsening health disparities."
Chirag Patel adds: "Socioeconomic status alone doesn't capture the full picture of health risk, and how we measure it matters. These findings set a clearer standard for how researchers should capture and report these variables so that predictive tools are accurate across the patients they're meant to serve."
In your coverage, please use this URL to provide access to the freely available article in PLOS One: https://plos.io/4xWqDdc
Citation: Cromer SJ, Gervis JE, Burnett-Bowie S-AM, Patel CJ (2026) Heterogeneous associations of socioeconomic status with metabolic disease in racial and ethnic subgroups in the United States: A cross-sectional cohort study in NHANES and All Of Us. PLoS One 21(7): e0351075. https://doi.org/10.1371/journal.pone.0351075
Author countries: USA.
Funding: SJC and JEG were supported by the American Diabetes Association (grant number 7-21-JDFM-005). CJP, SJC, and JEG were supported by the National Institute of Environmental Health Sciences (grant number R01ES032470) and National Institutes of Diabetes, Digestive, and Kidney Diseases (grant number R01DK137993). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
