Summarized & reviewed by The Peptide Dispatch Editorial Team · Last reviewed August 3, 2026
Almost every blood panel drawn in the United States includes a complete blood count, and almost every complete blood count reports a value called RDW, or red cell distribution width. It sits in the middle of the results, usually between the hematocrit and the platelet count, expressed as a percentage somewhere in the low teens. It costs nothing extra. It is already paid for. It has been printed…
This dispatch covers The Line on Your Blood Test That Predicts Mortality, and Why Nobody Reads It in the metabolic research category, authored by The Peptide Dispatch Editorial Team. Estimated reading time: 9 minutes. The Peptide Dispatch curates peer-reviewed peptide research for self-directed learners. All summaries are presented for Research Use Only and do not constitute medical advice.
Almost every blood panel drawn in the United States includes a complete blood count, and almost every complete blood count reports a value called RDW, or red cell distribution width. It sits in the middle of the results, usually between the hematocrit and the platelet count, expressed as a percentage somewhere in the low teens.
It costs nothing extra. It is already paid for. It has been printed on lab reports for decades.
And in most clinical encounters it is glanced at only when the hemoglobin is low, because that is what it was originally put there to do. If the hemoglobin is normal, the RDW is generally treated as noise.
The research literature has been saying something different for close to twenty years. Across very large general-population cohorts, RDW tracks all-cause mortality with a consistency that most purpose-built risk markers do not achieve. What follows is what the published evidence actually reports, including the substantial parts of it that remain unexplained.
Red blood cells are supposed to be uniform. A healthy bone marrow producing cells under stable conditions turns out a population of erythrocytes that are close to the same size as one another.
RDW quantifies how much they vary. It is a coefficient of variation of red cell volume, so a low value means the cells are tightly clustered around one size and a higher value means the population is spread out. The old clinical term for that spread is anisocytosis.
On most US lab reports the reference range runs roughly 11.5 to 14.5 percent, though the exact interval depends on the analyzer, so the printed range on your own report is the one that applies.
The reason it was added to the standard panel is diagnostic rather than prognostic. Iron deficiency produces small cells of uneven size, so it raises RDW early. Vitamin B12 and folate deficiency produce large cells. In a patient who is already anemic, RDW helps sort out which kind of anemia is present. That is a narrow and legitimate use, and it is essentially the whole of how the number gets used in routine practice.
The finding that changed the conversation came out of the Third National Health and Nutrition Examination Survey. Investigators followed 15,852 adults and examined RDW against mortality, published in Archives of Internal Medicine (DOI).
Estimated mortality rates rose roughly fivefold from the lowest to the highest quintile of RDW after accounting for age, and remained roughly twofold higher after full multivariable adjustment. A single standard deviation increment, which in that cohort was 0.98 percent, carried a 23 percent greater risk of all-cause death (HR 1.23, 95% CI 1.18 to 1.28).
The pattern was not confined to one disease. Risk of cardiovascular death rose (HR 1.22), as did cancer death (HR 1.28) and death from chronic lower respiratory disease (HR 1.32). A marker that predicts three unrelated causes of death is behaving less like a specific disease signal and more like a general readout of physiological condition.
The scale of the replication is what makes this hard to dismiss. A retrospective cohort in Alberta, Canada included every adult in the province with at least one RDW measurement between 2003 and 2016, which came to 3,156,863 people, and followed them a median of 6.8 years (DOI). Compared with people in the 25th to 75th percentiles, adjusted risk of death was higher in the 75th to 95th percentiles (HR 1.42), higher again in the 95th to 99th (HR 1.86), and highest above the 99th percentile (HR 2.18). The same graded pattern appeared for first myocardial infarction, first stroke or transient ischemic attack, incident cancer, hospitalization, and placement into long-term care.
Two points from that study deserve attention. RDW was not associated with progression to end-stage renal disease, which is a useful reminder that the marker is not simply flagging every bad outcome indiscriminately. And the related parameter SD-RBC, which reports the spread in femtoliters rather than as a coefficient of variation, performed slightly better than RDW across most endpoints. The authors suggest SD-RBC may be the superior of the two.
Longer follow-up gives the same answer. The Malmö Diet and Cancer cohort measured RDW in 27,063 participants aged 45 to 73 and followed them nearly twenty years, recording 9,388 deaths (DOI). Highest versus lowest quartile carried a 34 percent higher all-cause mortality risk, with cancer, cardiovascular and respiratory mortality all elevated.
That study also did something the others mostly skipped, which was to ask how much predictive value RDW actually adds. Adding it to a model already adjusted for age and sex moved the C-statistic from 0.732 to 0.737. Statistically significant, and honestly, small. This is a real marker with a modest incremental contribution, and anyone selling it as a crystal ball is overstating it.
If RDW were simply detecting undiagnosed iron deficiency, the association would collapse once you account for anemia. It does not.
A community cohort in Taiwan followed 3,226 adults for a median of 15.9 years and specifically stratified by anemia status (DOI). People with high RDW and no anemia still carried elevated risk of all-cause and non-cardiovascular death compared with people who had low RDW and no anemia. The authors concluded that RDW may precede anemia in predicting risk.
A prospective study in 1,479 hip fracture patients found the same thing more sharply. The association between admission RDW and long-term mortality was stronger in the subgroup without anemia than in the cohort overall (DOI).
And in a cohort of 109,675 maintenance hemodialysis patients, RDW outperformed hemoglobin, ferritin and iron saturation as a mortality predictor (DOI). In a population where iron status is measured obsessively and treated aggressively, the crude measure of cell size variability carried more prognostic information than the iron markers themselves.
So a normal hemoglobin does not neutralize a high RDW. The two are answering different questions.
There is no single accepted mechanism, and it would be dishonest to present one as settled.
The plausible account is that erythropoiesis is a demanding, continuous, high-throughput process. The marrow turns over roughly two million red cells per second, and doing that uniformly requires adequate iron, B12 and folate, functioning kidneys producing erythropoietin, an intact marrow, and an internal environment that is not chronically inflamed or oxidatively stressed. Any of those going wrong widens the size distribution of the output. Shortened or variable red cell survival in circulation widens it further.
Under that reading, RDW is not measuring a disease. It is measuring whether one of the body's most quality-controlled manufacturing processes is still running to specification, which is why it moves in inflammation, in nutritional depletion, in kidney disease, in liver disease and in malignancy alike. That non-specificity is simultaneously the reason it predicts so many outcomes and the reason it cannot tell you which one.
A 2024 analysis in JAMA Network Open pushed on this by combining RDW with serum albumin, another broad marker of nutritional and inflammatory status, as a simple ratio (DOI). Across 50,622 NHANES participants and 418,950 UK Biobank participants, elevated ratio was associated with substantially higher all-cause mortality (HR 1.83 and 2.08 respectively), with consistent elevations for cancer, heart disease, cerebrovascular, respiratory and diabetes mortality in both cohorts. Two cheap and universally available numbers, combined, separated risk more sharply than either alone.
This is where the honest boundary sits, and it matters more than the hazard ratios.
Every study above is observational. No trial has demonstrated that lowering RDW improves survival, and there is no reason to assume it would, because RDW is far more likely to be a consequence of underlying physiology than a cause of anything. Treating the number rather than what produced it would be a category error.
RDW also has no diagnostic specificity whatsoever. An elevated value tells you that something is perturbing red cell production or survival. It does not tell you what. It is a prompt to look, not an answer.
There is no established optimal target. The cohort data describe risk gradients across percentiles in populations, which is not the same as a validated personal threshold, and the incremental predictive gain in the Malmö analysis was genuinely modest.
The practical version is short.
Find the RDW on your most recent complete blood count. It is already there. Note whether it sits in the lower or upper part of your lab's printed range, rather than treating anything inside the range as equivalent, since the population data show a gradient that runs well inside conventional normal limits.
If it is drifting upward across serial panels while your hemoglobin stays normal, that trend is the more informative observation, and it is the kind of thing a single snapshot cannot show. Alongside it, the standard first questions are the unglamorous ones: iron studies including ferritin and transferrin saturation, B12 and folate, kidney and liver function, and a marker of inflammation.
None of this requires a specialized panel or a new test. It requires reading a line that was already run, already paid for, and already sitting in the chart.
The Peptide Dispatch publishes research summaries for educational purposes. This material describes what the published literature reports about laboratory markers and physiology. It is not medical advice, it is not a diagnosis, and it does not recommend or endorse any treatment, product, or provider. Laboratory results require interpretation by a licensed clinician in the context of your individual history. Always discuss testing and treatment decisions with a qualified healthcare professional.
Research summaries in this article are based on articles retrieved from PubMed.
Educational content — not medical advice. Effects described are drawn from cited research in study subjects.