RDW — red cell distribution width — measures how much your red blood cells vary in size. It sits on every complete blood count, it is almost never discussed, and it is the number on a standard panel with the most surprising relationship to long-term outcomes. Across a large body of observational research, a higher RDW is associated with higher all-cause mortality even in people who are not anaemic and whose RDW never leaves the normal range.
The Verdict
What RDW physically measures
A haematology analyser sizes red cells one at a time as they pass a sensor, building a distribution of tens of thousands of measurements. MCV reports the centre of that distribution. RDW reports its width — how spread out the sizes are around the average.
The same analyser applies the same logic to platelets on the same sample, which is where MPV comes from — the average platelet size, reported for the reason RDW is: a shift in the size distribution says something about production that the count alone does not.
That distinction is the whole value of the test. A uniform population of average cells and a mixture of small and large cells can share an identical MCV. Only RDW tells them apart. Because a widening distribution appears as soon as the marrow starts producing cells that differ from the ones already circulating, RDW frequently moves earlier than either MCV or haemoglobin.
One practical caveat: most labs report RDW-CV as a percentage, but some report RDW-SD in femtolitres, commonly quoted around 39–46 fL though varying more between analysers than RDW-CV does. They are different measurements and cannot be compared to each other. Check which your report uses, and read it against the range printed on it.
Standard ranges and what each band suggests
| Band | RDW-CV | What it suggests |
|---|---|---|
| Typical reference range | Roughly 11.5–14.5% | The band nearly every lab reports as normal (RDW-CV). Cutoffs vary slightly by analyser. |
| Low-normal | Under 13.5% | Uniform red cell population. This is where the observational risk curves sit at their lowest. |
| High-normal | 13.5–14.5% | Still unflagged. Most of the excess risk seen in cohort studies sits in and just above this band. |
| Raised | Above 14.5% | A meaningfully mixed red cell population. Worth explaining rather than repeating. |
| Markedly raised | Above 16% | Usually reflects an active nutritional deficiency, haemolysis, or marrow disorder. Needs assessment. |
What raises RDW
| Cause | How often it explains a result | Notes |
|---|---|---|
| Iron deficiency | The most common cause | RDW often rises before MCV falls, making it an early signal rather than a confirmatory one. |
| B12 or folate deficiency | Common | New cells are produced larger while older normal cells remain, widening the distribution. |
| Mixed deficiency | Frequently missed | Iron plus B12 or folate together. MCV can look normal while RDW is clearly raised. |
| Recent blood loss or haemolysis | Situational | A burst of young, larger cells widens the spread. Here a high RDW reflects a marrow responding properly. |
| Chronic inflammation | Common in older adults | Impairs iron handling and red cell production. Read alongside hsCRP. |
| Liver disease and chronic kidney disease | Common | Both alter red cell membranes and survival, independent of any vitamin deficiency. |
| Recent transfusion | Situational and easily forgotten | Two donor populations in one circulation widen the distribution mechanically. |
| Thalassaemia trait | Usually NOT a cause | Cells are small but uniformly small, so RDW is typically normal. This is what separates it from iron deficiency. |
The mortality association, stated carefully
This is the part of RDW worth understanding properly, because it is easy to overstate. In large cohort studies across general populations, cardiovascular disease, heart failure, kidney disease, and critical illness, higher RDW has been repeatedly associated with higher all-cause and cardiovascular mortality. The association persists after adjusting for haemoglobin, age, and conventional risk factors, and it is present in people whose RDW sits within the normal reference band.
The mechanism is not established. The most credible explanation is that RDW is not doing anything itself, but summarising several processes that each shorten life: chronic inflammation, oxidative stress, impaired iron handling, poor nutritional status, and reduced kidney and marrow function. Red cell production is sensitive to all of these, so the width of the size distribution ends up acting as a general readout of how well that machinery is working.
Two consequences follow, and both matter. First, RDW is a reasonable thing to notice on a panel you already have — it costs nothing extra and it is already printed. Second, and more importantly, lowering RDW is not a goal. This is association from observational data, not a demonstrated causal pathway, and no intervention has been shown to improve outcomes by reducing RDW. If a deficiency is driving it, correct the deficiency because the deficiency is worth correcting. Chasing the number itself has no evidence behind it.
Why the normal band hides the signal
A reference range reports the central 95% of a reference population, which answers whether a result is common rather than whether it is good. Normal vs optimal sets out the framework. RDW is one of the clearer examples: much of the excess risk observed in cohort studies sits between roughly 13.5% and the upper limit of normal — a band in which no result is ever flagged and no clinician is ever prompted.
What RDW means in combination
- High RDW + low MCV: the classic iron-deficiency pattern. Check ferritin and transferrin saturation, and look for a source of loss. TIBC is the measure that stays interpretable when inflammation has pushed ferritin up and hidden the deficiency.
- Normal RDW + low MCV: points towards thalassaemia trait rather than iron deficiency — the cells are uniformly small. Iron is not the answer.
- High RDW + high MCV: B12 or folate deficiency, alcohol, or a reticulocyte response after bleeding or haemolysis.
- High RDW + normal MCV: either an early single deficiency or two opposing deficiencies cancelling out in the average. Do not read the normal MCV as reassurance.
- High RDW + raised hsCRP: the inflammatory pattern. Here RDW is tracking inflammation rather than nutrition, and the inflammation is the thing to explain.
- High RDW + normal haemoglobin: the early-warning pattern, and the one most often ignored.
How and when to test
- No fasting or special preparation. RDW comes with any complete blood count at no additional cost.
- Note whether your lab reports RDW-CV or RDW-SD. Mixing the two across labs produces meaningless comparisons.
- Interpret it beside MCV, never on its own. The pair is informative; either alone is much less so.
- Mention any transfusion in the previous three months, which widens the distribution mechanically and has no bearing on your own marrow.
- Re-test at three months after correcting a deficiency, since the red cell population turns over across roughly 120 days.
- Keep your own series. A rise from 12.4% to 14.2% is entirely inside the normal band and is the kind of change worth explaining.
When a result warrants seeing a physician
- RDW above 16%, particularly alongside any abnormality in haemoglobin or MCV.
- A raised RDW with normal ferritin, B12, folate, and kidney function — an unexplained result that deserves a haematology opinion rather than a supplement.
- A rising RDW alongside falling haemoglobin, unexplained fatigue, breathlessness, or weight loss.
- Any raised RDW in someone with known kidney disease, liver disease, or heart failure, where it more often reflects the underlying condition.
- A raised RDW in an adult without an obvious cause, since identifying a source of occult blood loss matters more than correcting the index.
RDW is a pointer, not a diagnosis. It is unusually good at indicating that something in red cell production has changed, and it never identifies what — that judgement belongs with a clinician who can see the whole panel and your history.
Frequently Asked Questions
What is RDW in a blood test?
RDW stands for red cell distribution width. It measures how much your red blood cells vary in size — the technical term is anisocytosis. Where MCV reports the average size of your red cells, RDW reports the spread around that average. Most labs report RDW-CV as a percentage, with a normal range of roughly 11.5 to 14.5%. A high RDW means you have a mixture of noticeably small and noticeably large cells in circulation.
What does a high RDW mean?
It means your red blood cells are unusually varied in size, which happens whenever a new population of cells is being produced differently from the existing one. Iron deficiency is the most common cause, followed by B12 or folate deficiency, recent blood loss or haemolysis, chronic inflammation, and liver or kidney disease. A raised RDW is a prompt to look at the rest of the panel — it identifies that something has changed in red cell production without naming what.
Why is RDW associated with mortality?
This is one of the more striking findings in routine laboratory medicine, and it is genuinely well replicated: across many large cohorts, a higher RDW is associated with higher all-cause and cardiovascular mortality — including in people who are not anaemic and whose RDW sits inside the normal range. The honest answer to why is that nobody is certain. The leading explanation is that RDW is a summary marker of several things that independently shorten life: chronic inflammation, oxidative stress, poor nutritional status, and impaired kidney and marrow function. On that reading RDW is a symptom of ageing biology rather than a cause of it, which matters because it means lowering the number is not itself the goal.
What is a normal RDW level?
Roughly 11.5 to 14.5% for RDW-CV in adults, with modest variation between analysers. Some labs also report RDW-SD in femtolitres, with a different range — commonly quoted as roughly 39 to 46 fL, though published RDW-SD ranges vary considerably more between analysers than RDW-CV does. Check which measure your report uses, and use the range printed on it, because the two are not interchangeable.
Can RDW be high with normal haemoglobin?
Yes, and it is one of the more useful patterns on a blood count. RDW frequently rises before haemoglobin falls and before MCV changes, because the distribution widens as soon as a new population of differently-sized cells starts appearing. A raised RDW with a normal haemoglobin often means an early iron or B-vitamin deficiency that has not yet produced anaemia — early enough that correcting it is straightforward.
How do I lower my RDW?
By finding and correcting whatever is widening the distribution, not by targeting the number itself. If the cause is iron, B12, or folate deficiency, correcting the deficiency brings RDW down over two to three months as the red cell population becomes uniform again. If the driver is chronic inflammation, kidney disease, or liver disease, RDW tracks that condition and will follow it rather than lead. Treating RDW as a target in its own right is a mistake — there is no intervention that lowers it directly, and no evidence that doing so would help if there were.
How does RDW distinguish iron deficiency from thalassaemia trait?
Both produce small red cells, so MCV alone cannot separate them, and the distinction matters because the responses are opposite. In iron deficiency the marrow is producing progressively smaller cells while older normal-sized ones remain, so the distribution widens and RDW rises. In thalassaemia trait the cells have been uniformly small for life, so the distribution stays narrow and RDW is typically normal. The red cell count helps too — it is usually low in iron deficiency and normal or high in thalassaemia trait. Ferritin confirms.