The study, published in Frontiers in Epidemiology, drew on the National Health and Nutrition Examination Survey, or NHANES, a nationally representative program that combines household interviews, physical examinations, and laboratory testing. The researchers, led by Hui Ma of the Fifth People’s Hospital of Kunshan in Suzhou, China, identified 92,062 adults in the NHANES database from 1999 to 2016. After excluding participants younger than 20 and those missing key data, 17,412 low-income adults remained in the final analysis. Low-income status was defined as a family income-to-poverty ratio below 1, meaning household income fell under the federal poverty threshold. Deaths were tracked through linkage with the National Death Index through the end of 2016.

RDW is reported as the coefficient of variation of red blood cell volume, expressed as a percentage. In a healthy person, red cells are remarkably uniform in size, and RDW values typically hover around 12 to 13 percent. When the bone marrow produces red cells under stress, whether from iron deficiency, chronic inflammation, oxidative damage, or premature destruction of circulating cells, the resulting population of red cells becomes more heterogeneous, and RDW rises. Because the test is part of a standard complete blood count, it costs nothing extra to obtain, which is precisely why the researchers argue it deserves attention in settings where advanced diagnostics are scarce.

The statistical approach was deliberately layered. The team first used generalized additive models to explore whether the relationship between RDW and mortality was linear or curved. They then built three progressively adjusted Cox proportional hazards models: an unadjusted model, a model adjusted for sociodemographic factors such as age, sex, education, and race, and a fully adjusted model that also accounted for body mass index, smoking, alcohol consumption, histories of cancer, stroke, liver disease, hypertension, diabetes, and kidney disease, as well as other red cell indices including mean corpuscular volume, mean corpuscular hemoglobin, hemoglobin concentration, and hemoglobin itself. All analyses incorporated NHANES sampling weights to preserve national representativeness, and the proportional hazards assumption was verified using Schoenfeld residuals.

The headline finding was consistent across all three models. In the fully adjusted analysis, each 1 percent increase in RDW was associated with an 11 percent increase in the risk of all-cause mortality, with a hazard ratio of 1.11 and a 95 percent confidence interval of 1.08 to 1.15, a result that remained statistically significant at P less than 0.0001. When participants were sorted into quartiles of RDW, those in the highest quartile faced nearly double the mortality risk of those in the lowest, with a hazard ratio of 1.88 after full adjustment. The trend across quartiles was highly significant, and the association held even though the cohort carried a heavy burden of pre-existing disease, which is typical of low-income populations.

The most intriguing result, however, was the shape of the dose-response curve. Rather than rising steadily, mortality risk climbed steeply as RDW increased, then flattened out. Piecewise regression analysis pinpointed a turning point at an RDW of 15.9 percent. Below that threshold, every 1 percent increase in RDW was associated with a 29 percent higher risk of death, with a hazard ratio of 1.29. Above 15.9 percent, the association disappeared entirely, with a hazard ratio of 1.00 that was not statistically significant. A log-likelihood ratio test confirmed that the two-segment model fit the data significantly better than a single straight line.

The authors offer two possible explanations for this saturation phenomenon. One is biological: the harmful processes that elevate RDW, including chronic inflammation, ineffective erythropoiesis, and oxidative stress, may reach a plateau, beyond which additional variability in red cell size simply adds no new prognostic information. The other is statistical: people with markedly elevated RDW often have severe underlying conditions such as advanced malignancy, end-stage renal disease, or myelodysplastic syndromes, in which death is driven by the primary disease rather than by RDW itself. The researchers also note that relatively few participants had RDW values above 15.9 percent, which may have limited statistical power in that range. Importantly, they emphasize that the 15.9 percent cutoff is exploratory and data-driven, derived from the software’s search for the optimal turning point, and has not been validated in independent cohorts.

Why would RDW be especially informative in poor communities? The authors argue that socioeconomic disadvantage acts on red cell biology through multiple converging pathways. Low-income adults are more likely to experience deficiencies in iron, folate, and vitamin B12, all essential for producing healthy red cells. They face greater exposure to environmental stressors, higher rates of chronic inflammatory conditions, and reduced access to preventive and medical care. Together these forces disrupt erythropoiesis and red cell homeostasis, producing the size variability that RDW captures. In this sense, RDW may function as an integrative biomarker, reflecting not just hematological abnormalities but the cumulative physiological toll of sustained poverty. The baseline data support this picture: higher RDW quartiles were associated with older age, higher body mass index, lower hemoglobin, and greater prevalence of cancer, stroke, kidney disease, hypertension, and diabetes.

The findings align with a growing body of literature linking elevated RDW to poor outcomes in specific clinical settings, including sepsis, traumatic brain injury, hip fracture, intracerebral hemorrhage, acute aortic dissection, and various cancers. What distinguishes the new study is its focus on a socioeconomically defined subgroup within a general population sample rather than hospitalized patients, and the authors report that this is the first investigation of the RDW-mortality association specifically in low-income adults. Still, they are careful about interpretation. The hazard ratio of 1.11 is modest, and in an observational study such an effect may partly reflect residual confounding by unmeasured factors such as inflammatory markers, iron status, or subclinical disease. The authors explicitly caution that baseline RDW should not be treated as an established clinical prediction tool, and that the observed association does not imply a causal effect of RDW itself on mortality.

There are also limitations worth weighing. The complete-case analysis excluded 74,650 individuals, about 81 percent of the initial sample, mostly because of missing covariate data, a rate high enough to raise concerns about selection bias even with sampling weights applied. The authors counter that the association remained stable and significant across all adjusted models, and that the biological plausibility of the RDW-mortality link is well documented elsewhere. They also acknowledge that a formal competing-risks analysis, such as a Fine-Gray model, was not performed, and that future work should disentangle cause-specific mortality in this population. The practical takeaway, they suggest, is not that RDW can predict an individual’s fate with precision, but that this free, routinely available number could serve as an accessible population-level signal in resource-limited settings, flagging patients who warrant closer monitoring or further evaluation. Before that happens, the threshold effect and the overall association will need confirmation in independent, prospectively designed cohorts.

Subject of Research: Association between red blood cell distribution width and all-cause mortality in low-income adults

Article Title: Red blood cell distribution width and all-cause mortality in low-income populations: a NHANES-based cohort study

Article References: Ma, H., Li, Y.-H., Zhao, J.-D., Gu, Y.-W., & Zhu, R.-T. (2026). Red blood cell distribution width and all-cause mortality in low-income populations: a NHANES-based cohort study. Frontiers in Epidemiology, 6, Article 1746886. https://doi.org/10.3389/fepid.2026.1746886

Image Credits: AI Generated

DOI: 10.3389/fepid.2026.1746886

Keywords: red blood cell distribution width, all-cause mortality, NHANES, low-income populations, cohort study, Cox regression, anemia, health disparities, biomarker, epidemiology, poverty, public health