One Inflammatory Molecule May Signal Cognitive Impairment Across Diverse Aging Populations

A single molecule circulating in the blood may carry the same warning about brain health for older adults regardless of race or ethnicity. That is the striking implication of a new analysis from one of the largest multi-ethnic aging studies in the United States, which found that elevated levels of the inflammatory protein interleukin-6, or IL-6, were consistently associated with cognitive impairment among Black, Hispanic, and non-Hispanic White older adults. The findings, drawn from the Health and Aging Brain Study–Health Disparities, or HABS-HD, cohort, suggest that IL-6 could serve as a broadly generalizable biomarker of dementia risk, a possibility with major consequences for equitable detection and prevention of Alzheimer’s disease and related dementias.

The research team, led by investigators at the Tri-institutional Center for Translational Research in Neuroimaging and Data Science and Georgia State University, analyzed complete data from 2,187 older adults, including 963 non-Hispanic White, 277 Black, and 947 Hispanic participants. Cognitive impairment was defined as a diagnosis of either mild cognitive impairment or dementia, compared against cognitively unimpaired status. Using logistic regression, the researchers tested whether higher concentrations of plasma IL-6, measured with ultrasensitive Simoa assays and log-transformed to correct for the right-skewed distribution typical of circulating cytokines, predicted greater odds of impairment after adjusting for age, education, sex, depressive symptoms, ethnoracial group, and APOE ε4 carrier status, the strongest known genetic risk factor for late-onset Alzheimer’s disease.

The results were unambiguous. Each tenfold increase in plasma IL-6 was associated with 87 percent greater odds of cognitive impairment in the full sample, with an odds ratio of 1.87 and a 95 percent confidence interval of 1.31 to 2.68, a statistically robust effect. Critically, formal interaction testing revealed no significant difference in this association across ethnoracial groups, with the likelihood-ratio test yielding a p-value of 0.172. In other words, the inflammatory signal appeared to operate similarly regardless of group membership, pointing toward a shared biological pathway rather than one specific to any single population.

Because demographic imbalance between groups can distort associations in multi-ethnic cohorts, the team conducted rigorous sensitivity analyses using propensity score matching. They created 1:1 matched subsamples in which White participants were matched exactly to Black or Hispanic participants on sex and cognitive status, and balanced on age and education. The Black/White matched sample included 554 participants and the Hispanic/White matched sample included 794, later reduced to 788 for adjusted analyses because six participants lacked APOE ε4 data. Within these carefully balanced subsamples, the IL-6 association not only persisted but strengthened, with odds ratios of 2.62 in the Black/White matched sample and 2.24 in the Hispanic/White matched sample, both statistically significant, and again with no significant IL-6-by-ethnoracial-group interactions.

The study also examined depressive symptoms, measured with the 15-item Geriatric Depression Scale, which have long been recognized as a robust predictor of late-life cognitive decline. Meta-analytic evidence indicates that late-life depression is associated with roughly 1.85-fold higher risk of all-cause dementia. In this cohort, each additional point on the depression scale was independently associated with approximately 6 percent greater odds of cognitive impairment, an effect that held in both the full sample and the matched subsamples. The convergence of depression and inflammation as independent correlates of impairment raises the question of whether one might statistically transduce the other.

Mediation analyses provided modest but intriguing support for that hypothesis. Higher depression scores were significantly associated with higher IL-6 concentrations, and IL-6 remained significantly associated with cognitive impairment after adjustment for depressive symptoms. The indirect effect of depression on cognitive impairment through IL-6 was statistically significant in the full sample, with a Sobel test p-value of 0.014, though the researchers emphasize that IL-6 accounted for only a small proportion of the total association. In the Hispanic/White matched sample, the indirect effect approached significance, while in the Black/White matched sample the depression-to-IL-6 path was not significant, resulting in a nonsignificant indirect effect. These patterns suggest possible ethnoracial variation in how inflammation links mood to cognition, though the authors caution that subgroup differences in statistical significance do not establish significant between-group differences.

Secondary analyses of the broader cytokine panel revealed a more selective inflammatory profile than a generalized elevation across all immune markers. Plasma IL-10 and TNF-α showed no significant associations with cognitive impairment in fully adjusted models, despite their frequent discussion in the Alzheimer’s neuroinflammation literature. Notably, IL-5 showed an inverse association, such that higher concentrations corresponded to lower odds of impairment, an exploratory signal that may reflect a distinct type 2 immune-regulatory pathway whose relationship to cognition differs from canonical pro-inflammatory markers. The selectivity of the IL-6 finding strengthens the interpretation that this particular cytokine captures a chronic inflammatory pathway most relevant to cognitive aging in this cohort.

Mechanistically, IL-6 occupies a unique position at the intersection of immune signaling and neurodegeneration. The cytokine coordinates both innate and adaptive immune responses, and while transient IL-6 activity can be protective, sustained IL-6 trans-signaling is predominantly proinflammatory and can perpetuate chronic immune activation. In the brain, IL-6 signaling promotes activation of microglia and astrocytes, the resident immune cells that release further pro-inflammatory mediators capable of impairing synaptic function and accelerating neuronal loss. IL-6 has been linked to amyloid-β deposition, tau hyperphosphorylation, reduced hippocampal and cortical volumes, and poorer cognitive performance in prior research. Psychological stress may upregulate IL-6 through repeated activation of the sympathetic nervous system and the hypothalamic–pituitary–adrenal axis, and over time altered β-adrenergic signaling and reduced glucocorticoid sensitivity can weaken normal anti-inflammatory feedback, providing a plausible route by which chronic stress and depression translate into systemic inflammation.

The researchers also explored whether chronic stress might account for the IL-6 association, given that social determinants of health such as discrimination, ongoing stressors, and inequities in healthcare access are linked to Alzheimer’s risk. In an exploratory sensitivity analysis with data from 2,168 participants, higher chronic stress was associated with higher IL-6 concentrations, but chronic stress was not independently associated with cognitive impairment after adjustment for IL-6 and remaining covariates. The IL-6 association remained significant and materially unchanged when chronic stress was added to the model, suggesting that stress did not explain away the inflammatory signal. This finding points toward IL-6 as a potential convergence point where multiple upstream exposures, from vascular and metabolic burden to psychosocial adversity, may jointly shape vulnerability to cognitive decline.

The implications for equitable biomarker development are significant. Black older adults face approximately twice the risk, and Hispanic older adults about 1.5 times the risk, of developing Alzheimer’s disease and related dementias compared with non-Hispanic White adults, yet these groups remain persistently underrepresented in dementia research. If interventions targeting IL-6, whether through lifestyle modification, anti-inflammatory therapies, or depression management, could benefit diverse populations without requiring race or ethnicity-specific tailoring, the pathway toward precision prevention becomes more scalable and more just. The authors emphasize, however, that the cross-sectional design of this study precludes conclusions about temporal ordering, and longitudinal analyses within HABS-HD and other diverse cohorts will be needed to determine whether baseline IL-6 predicts incident cognitive decline, whether changes in IL-6 mediate treatment effects on cognition, and whether IL-6 represents a causal contributor to neurodegeneration or a downstream marker of an already active disease process. As the scientific community increasingly recognizes brain health as a modifiable construct, IL-6 shows promise as a scalable, generalizable biomarker for identifying midlife adults with inflammation-related concerns about cognitive aging, and future work will be critical for translating these biomarker associations into actionable strategies to reduce dementia burden across all racial and ethnic groups.

Subject of Research: Plasma IL-6 as a shared inflammatory marker of cognitive impairment across racially and ethnically diverse older adults: matched analyses from the HABS-HD study

Article Title: Plasma IL-6 as a shared inflammatory marker of cognitive impairment across racially and ethnically diverse older adults: matched analyses from the HABS-HD study

Article References: Misiura, M. B., Joseph, T. O., Cooper, J. A., Champion, G., Burnette, J., Bartlett, A., Thatcher, K., Austin, N. C., & Dotson, V. M. (2026). Plasma IL-6 as a shared inflammatory marker of cognitive impairment across racially and ethnically diverse older adults: matched analyses from the HABS-HD study. Frontiers in Dementia, 5, Article 1930196. https://doi.org/10.3389/frdem.2026.1930196

Image Credits: AI Generated

DOI: 10.3389/frdem.2026.1930196

Keywords: Plasma, IL-6, shared, inflammatory, marker, cognitive, impairment, across, racially, ethnically, diverse, older