If your hemoglobin came back at 12.4 g/dL as a man, or 12.1 g/dL as a woman, most lab reports print it in black rather than red. It sits inside the reference range, so nothing gets flagged. New research published on 8 September 2026 suggests that reassurance may be a little too quick.
A University of Glasgow team followed close to half a million UK Biobank participants for more than a decade and found that people whose hemoglobin sat just above the anemia threshold had a higher risk of death than people further up the same normal range. Nothing about this changes a diagnosis on its own, and it is an observational finding rather than proof. But it does change how a borderline number deserves to be read.
新しい研究が実際に明らかにしたこと
The researchers compared hemoglobin concentration with deaths from any cause, from cardiovascular disease and from cancer, over a median follow-up of more than ten years. The shape of the relationship was a U: risk rose at both the low end and the high end, and bottomed out somewhere in between.
The interesting part is where that bottom sat. It was not immediately above the anemia cut-off. It was 1 to 3 g/dL above it, in both men and women. People with what the authors call borderline anemia, meaning a value just clearing the threshold, carried more risk than people sitting in the upper part of the normal band.
The team is careful about what this does and does not mean. An association is not a cause. A borderline hemoglobin may simply be a marker that something else is going on, rather than a problem in itself. The authors explicitly say the findings do not justify redrawing the anemia thresholds today, and that whether investigating these people improves their outcomes still has to be tested directly.
Where the anemia threshold comes from
Anemia is currently defined by World Health Organization cut-offs that have barely moved in fifty years: hemoglobin below 13 g/dL in men and below 12 g/dL in women. Everything above those numbers is reported as normal.
A single line has obvious advantages. It is simple, it is the same everywhere, and it tells a busy clinician when to act. The limitation is equally obvious: biology does not change at a decimal point. Two people separated by 0.2 g/dL get very different reports, and one of them is told nothing is wrong. That is the tension this study puts a number on.
The three zones on a hemoglobin result
Reading the result as three zones rather than two is closer to what the data show.
| Zone | Hemoglobin (men / women) | How to read it |
|---|---|---|
| Below the threshold | Under 13 / under 12 g/dL | Anemia by the current definition. Warrants a cause being looked for. |
| 境界域 | Roughly 13 to 14 / 12 to 13 g/dL | Reported as normal. The zone where risk already started to rise in this study. |
| Comfortably normal | Roughly 14 to 16 / 13 to 15 g/dL | Where the lowest mortality was observed in the cohort. |
| 高 | Above the upper reference limit | Uncommon, and also associated with higher risk. Needs medical assessment. |
Reference ranges differ slightly between laboratories, and altitude, pregnancy and age all shift them, so the numbers on your own report are the ones that count. For the wider picture of what this marker does, you can read our complete guide to hemoglobin levels.
Why a borderline result is often an iron story
One of the study authors made a practical point: borderline anemia comes with a high prevalence of iron deficiency. That matters, because iron runs down in stages, and hemoglobin is the last thing to fall.
Iron stores empty first. Red cells then start being built with less hemoglobin in them, which is why the mean corpuscular volume often drifts down before anything else. Only at the end does the hemoglobin itself drop below the line. Someone in the borderline zone may already be well into that sequence.
What the recent research adds
A 2025 review in JAMA on iron deficiency in adults set out how common this is: in high-income countries, roughly 38 percent of non-pregnant women of reproductive age have iron deficiency without anemia, compared with about 13 percent who have iron-deficiency anemia. In plain terms, for every woman whose blood test flags anemia, there are two or three whose iron is low while the hemoglobin still reads normal. The review also gives the number that identifies it: a ferritin below roughly 30 ng/mL, in the absence of inflammation.
A smaller 2023 hospital series in Blood looked specifically at people whose hemoglobin was above 12 g/dL. Their average hemoglobin was 13.4 g/dL, entirely unremarkable, yet their average ferritin was 16 ng/mL, and 83 percent of them reported fatigue. Most were treated with iron, and roughly six in ten said their symptoms improved, without their hemoglobin changing much.
What this changes for you is modest but concrete. A hemoglobin sitting near the bottom of its range, especially alongside tiredness, breathlessness on exertion, restless legs or hair and nail changes, is a reasonable prompt to ask whether iron stores have been measured. The same blood draw usually allows a ferritin blood test. These are small studies and a narrative review, not treatment recommendations, and iron supplements should never be started without a measured deficiency.
医師に相談すべきタイミング
A borderline hemoglobin on its own, in someone who feels well, is not an emergency and rarely needs urgent action. It is worth raising with a doctor when it comes with any of the following.
- Fatigue that is new, persistent, or out of proportion to your routine.
- Breathlessness climbing stairs, palpitations, or unusual pallor.
- Heavy periods, or any visible blood loss from the gut or urine.
- A hemoglobin that has been drifting down across successive tests, even inside the normal range.
- A known condition that affects red cell production, such as chronic kidney disease or an inflammatory disease.
Trend matters more than a single reading. A value of 13.2 g/dL means something different in someone who was at 15 two years ago than in someone who has always been there. The lab also reports your hematocrit, and the report shows the mean corpuscular volume, both of which help place a borderline number in context.
用語集
- Hemoglobin: the iron-containing protein inside red blood cells that carries oxygen from the lungs to the tissues.
- Anemia: too little hemoglobin to meet the body’s oxygen needs, defined by the WHO thresholds of 13 g/dL in men and 12 g/dL in women.
- Borderline anemia: a hemoglobin sitting just above those thresholds, reported as normal.
- Ferritin: the protein that stores iron. It falls before hemoglobin does, which makes it the earlier signal of iron deficiency.
- Cohort study: research that follows a large group over time and records what happens. It can show associations but cannot prove cause.
- U-shaped association: a relationship where risk is higher at both the low and high ends of a measurement, and lowest in the middle.
よくある質問
Does this mean the anemia threshold is wrong?
No. The authors are explicit that their findings do not justify changing the WHO thresholds. What they suggest is that a single cut-off cannot capture everything, and that the zone just above it deserves more research.
My hemoglobin is 12.5 and I am a woman. Should I be worried?
Worried, no. Curious, reasonably. That value is normal by the current definition. If you feel well and it has been stable, it is very likely nothing. If you are tired, have heavy periods, or it has been falling, it is worth mentioning at your next appointment.
Should I take iron supplements if my hemoglobin is borderline?
Not on the basis of a hemoglobin number alone. Iron overload causes its own harm, and iron should only be taken when a deficiency has been measured, typically on ferritin or transferrin saturation, and when the reason for it has been considered.
Does a borderline hemoglobin always mean iron deficiency?
No. Vitamin B12 or folate deficiency, chronic kidney disease, chronic inflammation and blood loss can all produce the same picture, and some people simply run at the lower end of normal. That is exactly why the cause is investigated rather than assumed.
How often should hemoglobin be rechecked?
There is no universal interval for someone who feels well. Where a doctor is following a borderline value or has started treatment, a recheck after a few weeks to a few months is usual. The timing is a clinical decision.
参考文献
- University of Glasgow — Borderline low haemoglobin levels are linked to a higher risk of death (8 September 2026)
- Annals of Internal Medicine — Association Between Blood Hemoglobin Concentration and Mortality by Age and Sex: A Prospective Cohort Study
- MedlinePlus (NIH / National Library of Medicine) — CBC blood test
- Auerbach M. et al. — Iron Deficiency in Adults: A Review, JAMA, 2025
- Sevgi M. et al. — Symptomatic Iron Deficiency without Anemia: An Underrecognized Phenomenon, Blood, 2023
関連記事
- ヘモグロビン値：血液検査の結果が意味すること
- Ferritin blood test: what your results mean
- 血清鉄検査：高値・低値・正常値の見方
- Hematocrit test: what your results mean and normal ranges
- MCV blood test: what it measures and what results mean
AI DiagMeで血液検査の結果をわかりやすく確認
A hemoglobin value means little on its own. Read alongside ferritin, MCV, hematocrit and the rest of the count, it starts to tell a story. AI DiagMe turns a full blood, urine or stool report into plain-language explanations, so you arrive at your appointment knowing which numbers are worth asking about. AI DiagMeで血液検査の結果をわかりやすく確認.
