Understanding Your Blood Test Results
A doctor explains what each line on a blood test report measures, what the numbers mean in both mg/dL and mmol/L, and why one odd result rarely matters.
The short version
- A reference range is not a health target. It is the middle 95% of a reference population, so roughly 1 in 20 healthy people sits outside it on any given test.
- Trends over time and the pattern across several results matter far more than a single value flagged in bold red.
- The tests that change management most often are hemoglobin with ferritin, HbA1c, kidney function, and thyroid function, not the long tail of extras on private panels.
- Slightly abnormal liver enzymes, a marginal white cell count, or a borderline TSH are usually repeated before anything is done, and often normalize on their own.
See a doctor promptly if
These are the signs that change this from something to read about into something to act on.
- Hemoglobin well below the normal range with breathlessness, chest pain, or dizziness on standing
- Potassium below 3.0 mmol/L or above 6.0 mmol/L. This needs same-day contact, not a routine appointment
- A sudden large rise in creatinine, or a new drop in eGFR, especially with reduced urine output
- Liver enzymes several times the upper limit with yellow eyes, dark urine, pale stools, or right upper abdominal pain
- Platelets under 50 x10^9/L, or a very high or very low white cell count with fever
Most people get their blood results as a PDF full of abbreviations, with two or three numbers in red, hours before anyone explains them. That gap is where the worry lives. What follows is a line-by-line translation of the tests that turn up on ordinary reports, what they are actually measuring, and how a doctor reads them.
One thing to hold on to from the start: a blood test is a photograph, not a verdict. It captures one morning in a body that changes hour by hour.
What a "normal range" actually is#
A reference range is not the boundary between health and disease. Laboratories build it by testing a large group of apparently healthy people and taking the middle 95% of their results. The 2.5% above and 2.5% below are still healthy people. They just sit at the edges of a bell curve.
This means that if you run twenty different tests on a completely healthy person, on average one will come back flagged. Order a forty-test private panel and two flags are the expected result, not a finding.
Ranges also differ between laboratories, because they depend on the analyser, the method and the local population. They differ by age, sex, pregnancy and sometimes ethnicity. Always read a result against the range printed on that same report.
The panels, line by line#
Most reports are built from a handful of standard groupings. Each one below is a separate request a doctor ticks, and each answers a different question.
Full blood count (CBC)#
The complete blood count measures the cells: oxygen carriers, infection fighters, and clotting fragments.
| Test | Typical adult range | What it means when abnormal |
|---|---|---|
| Hemoglobin (Hb) | Men 13.0-17.0 g/dL (130-170 g/L); women 12.0-15.0 g/dL (120-150 g/L) | Low = anemia. High = dehydration, smoking, sleep apnoea, altitude, rarely a marrow disorder |
| Haematocrit (Hct/PCV) | Men 40-52%; women 36-46% | Moves with hemoglobin; sensitive to hydration |
| MCV (average red cell size) | 80-100 fL | Low = iron deficiency or thalassaemia trait. High = B12 or folate deficiency, alcohol, thyroid, some medicines |
| MCH | 27-32 pg | Low alongside low MCV points to iron deficiency or a hemoglobin trait |
| RDW (variation in cell size) | 11.5-14.5% | High early in iron deficiency, because new small cells mix with older normal ones |
| White cell count (WBC) | 4.0-11.0 x10^9/L | High with infection, steroids, stress, smoking. Low with viral illness, some drugs, and in some healthy people of African ancestry |
| Neutrophils | 2.0-7.5 x10^9/L | The bacterial-infection workhorse; also rises with physical stress and steroids |
| Lymphocytes | 1.0-4.0 x10^9/L | Often high in viral illness; persistently high needs review |
| Eosinophils | 0.0-0.5 x10^9/L | Allergy, asthma, drug reactions, and parasitic infection. The last matters in South Asia and Africa |
| Platelets | 150-400 x10^9/L | High with inflammation, infection or iron deficiency. Low with viral illness, alcohol, some drugs, dengue |
Two practical notes. First, a mildly raised white cell count a few days after a cold is expected. Second, in dengue-endemic regions a falling platelet count with fever is followed closely rather than treated as an incidental finding.
Metabolic panel: kidneys, salts and sugar#
This group is often labeled U&E, BMP, CMP or "renal profile".
| Test | Conventional | SI | Notes |
|---|---|---|---|
| Sodium | 135-145 mEq/L | 135-145 mmol/L | Low is common with certain diuretics, some antidepressants, and excess plain water |
| Potassium | 3.5-5.0 mEq/L | 3.5-5.0 mmol/L | A falsely high result is common if the sample haemolysed in transit. This is why a repeat is often requested |
| Chloride | 98-107 mEq/L | 98-107 mmol/L | Interpreted with bicarbonate, not alone |
| Bicarbonate | 22-29 mEq/L | 22-29 mmol/L | Low in acidosis, kidney disease, prolonged diarrhea |
| Urea / BUN | BUN 7-20 mg/dL | Urea 2.5-7.1 mmol/L | Rises with dehydration and high protein intake, not only kidney disease |
| Creatinine | Men 0.7-1.3 mg/dL; women 0.6-1.1 mg/dL | Men 62-115 umol/L; women 53-97 umol/L | Depends on muscle mass. A bodybuilder and a frail 80-year-old with the same number do not have the same kidneys |
| eGFR | . | Over 90 mL/min/1.73m2 normal | Below 60 sustained for three months defines chronic kidney disease |
| Fasting glucose | Under 100 mg/dL normal | Under 5.6 mmol/L | 100-125 mg/dL (5.6-6.9 mmol/L) prediabetes; 126 mg/dL (7.0 mmol/L) or above suggests diabetes |
| Calcium (corrected) | 8.5-10.2 mg/dL | 2.12-2.55 mmol/L | Always interpret with albumin; high calcium needs a PTH check |
| Albumin | 3.5-5.0 g/dL | 35-50 g/L | Low with liver disease, malnutrition, inflammation, protein loss in urine |
Creatinine is worth dwelling on. It is a muscle breakdown product, so it reflects both kidney filtration and how much muscle you carry. A very muscular young man can have a creatinine at the top of the range with flawless kidneys. A small elderly woman can have a "normal" creatinine with meaningfully reduced kidney function. eGFR adjusts for age and sex, which is why it is the more useful number.
Liver tests#
"Liver function tests" is a misleading name. Most of the panel measures liver injury, not liver function. Only albumin, bilirubin and clotting time really speak to function.
| Test | Typical range | What a raised value suggests |
|---|---|---|
| ALT | 7-56 U/L (lab-dependent) | The most liver-specific enzyme. Commonly raised in fatty liver, alcohol, viral hepatitis, some medicines |
| AST | 10-40 U/L | Also found in muscle and heart. Hard exercise or a muscle injury can raise it |
| ALP | 44-147 U/L | Bile duct problems, some bone conditions, normal pregnancy, and higher in growing adolescents |
| GGT | Men 10-71 U/L; women 6-42 U/L | Sensitive to alcohol and to certain drugs; helps show whether a raised ALP came from liver or bone |
| Bilirubin (total) | Under 1.2 mg/dL | Under 21 umol/L. A mildly high unconjugated bilirubin with everything else normal is often Gilbert syndrome, a harmless inherited variant affecting up to 1 in 20 people |
| Albumin | 3.5-5.0 g/dL (35-50 g/L) | Falls in established liver disease and in chronic inflammation |
The most common cause of mildly raised ALT worldwide is now metabolic (fatty) liver disease, tied to weight, insulin resistance and triglycerides. The second most common in many clinics is alcohol. Both improve with the same measures, and a repeat test after two to three months of change is genuinely informative.
Lipids: cholesterol and triglycerides#
Cholesterol is not a poison; it is a building material carried around in particles. The tests count what is in those particles.
| Test | Conventional (mg/dL) | SI (mmol/L) | Comment |
|---|---|---|---|
| Total cholesterol | Desirable under 200 | Under 5.2 | Least useful number on the panel by itself |
| LDL cholesterol | Optimal under 100 | Under 2.6 | Targets are lower for people with existing heart disease or diabetes |
| HDL cholesterol | Men above 40; women above 50 | Above 1.0 / above 1.3 | Higher is generally better, but raising it with drugs has not improved outcomes |
| Triglycerides | Under 150 | Under 1.7 | Very responsive to alcohol, refined carbohydrate and recent meals |
| Non-HDL cholesterol | Under 130 | Under 3.4 | Total minus HDL; a better single risk marker than LDL alone |
| Lipoprotein(a) | Under 50 mg/dL (approx.) | Under 125 nmol/L | Largely genetic; worth measuring once, especially with early family heart disease |
A number on its own does not tell you your risk. Risk calculators combine cholesterol with age, sex, blood pressure, smoking and diabetes, and a South Asian background is a recognized risk multiplier in UK tools. A moderately raised LDL in a 35-year-old non-smoker with normal blood pressure means something very different from the same LDL in a 62-year-old smoker with diabetes.
HbA1c and blood sugar#
HbA1c measures the proportion of hemoglobin with glucose stuck to it. Because red cells live about three months, it gives an average of roughly the last 8-12 weeks. No fasting needed, no single bad breakfast to blame.
| Category | HbA1c (%) | HbA1c (mmol/mol) | Fasting glucose |
|---|---|---|---|
| Normal | Below 5.7 | Below 39 | Below 100 mg/dL (5.6 mmol/L) |
| Prediabetes | 5.7-6.4 | 39-47 | 100-125 mg/dL (5.6-6.9 mmol/L) |
| Diabetes | 6.5 or above | 48 or above | 126 mg/dL (7.0 mmol/L) or above |
HbA1c is unreliable when red cell lifespan is abnormal: iron deficiency anemia can push it up slightly, while haemolysis, recent blood loss, pregnancy and some hemoglobin variants can pull it down. In those situations a fasting glucose or an oral glucose tolerance test answers the question better. Diagnosis normally needs two abnormal results, or one abnormal result plus clear symptoms.
Thyroid function#
| Test | Typical range | Interpretation |
|---|---|---|
| TSH | 0.4-4.0 mIU/L | Rises when the thyroid is underactive, falls when it is overactive. It moves in the opposite direction to the hormone |
| Free T4 | 0.8-1.8 ng/dL (10-23 pmol/L) | Confirms whether the gland output is genuinely low or high |
| Free T3 | 2.3-4.2 pg/mL (3.5-6.5 pmol/L) | Added mainly when overactivity is suspected |
| TPO antibodies | Negative / low titre | Positive suggests autoimmune thyroid disease, which predicts future underactivity |
A mildly raised TSH with a normal free T4 is called subclinical hypothyroidism. It is common, it frequently returns to normal on a repeat test six to twelve weeks later, and it is not automatically treated. TSH also dips during any serious illness, so testing someone in hospital with pneumonia produces results that mean very little.
Iron studies#
| Test | Typical range | What it tells you |
|---|---|---|
| Ferritin | 15-300 ug/L (men), 15-200 ug/L (women) | The stores. Below 30 ug/L means iron deficiency in almost anyone; below 15 ug/L is definitive |
| Serum iron | 60-170 ug/dL (10-30 umol/L) | Swings hour to hour and with supplements: weak on its own |
| Transferrin / TIBC | TIBC 240-450 ug/dL | Rises in iron deficiency as the body reaches for more iron |
| Transferrin saturation | 20-50% | Under 20% supports deficiency; persistently above 45% raises the question of iron overload |
The trap with ferritin is that it is also an inflammatory protein. Infection, obesity, liver disease and chronic inflammation all raise it, which can mask deficiency. When inflammation is likely, a ferritin under 100 ug/L with a low transferrin saturation can still mean iron deficiency, and CRP is often checked alongside to interpret it.
Vitamin D and B12#
25-hydroxyvitamin D is the storage form and the one measured. Deficiency is usually defined as below 20 ng/mL (50 nmol/L), insufficiency as 20-30 ng/mL (50-75 nmol/L), and levels above 30 ng/mL (75 nmol/L) as sufficient. Above 100 ng/mL (250 nmol/L) is a zone where harm becomes possible.
Vitamin B12 below roughly 200 pg/mL (148 pmol/L) is low; 200-300 pg/mL (148-221 pmol/L) is a gray zone where a methylmalonic acid or homocysteine level clarifies things. Metformin, long-term acid-suppressing medicines and a diet with little animal food all lower B12 over years.
What "slightly out of range" usually means#
Before anyone acts on a marginal result, a good clinician runs through a short list.
- Was there a technical explanation? Haemolysed sample, a drip in the same arm, a delayed sample, the wrong tube.
- Was there a physiological explanation? Dehydration, a heavy gym session in the previous 48 hours, a recent infection, a very late meal, pregnancy, the menstrual cycle.
- Is it consistent with anything the person actually feels?
- What did the previous result show?
- Does the pattern hang together? Iron deficiency should show a low ferritin, a low MCV, a high RDW and often a high platelet count. A low ferritin with everything else pristine may simply be an early store problem.
The answer to a single mildly abnormal result is very often "repeat it in six to twelve weeks and look at the direction of travel". That is not fobbing you off. It is the most accurate thing available.
What I actually see in clinic#
The pattern I see most is a person who has paid for a broad panel, arrives holding four pages, and is frightened by three red flags, an ALT of 48, an eosinophil count a fraction high, and a vitamin D of 24 ng/mL, while the finding that explains their exhaustion, a ferritin of 11 ug/L, is printed in black because it sits just inside the printed range.
The second pattern is the one that quietly matters: a person whose every result is technically normal, but whose HbA1c has moved 5.2, 5.5, 5.8 across three years. Nothing was ever flagged. Everything was heading somewhere. The trend inside the normal range is often the most valuable information on the report, and it is the part almost nobody looks at.
I also see a lot of anxiety created by a single number sent by text with no context. If a result arrives and you cannot tell whether it matters, that is not a failure of intelligence. It is a failure of how the result was delivered.
At work#
Occupational medicine adds a layer most people do not expect. In pre-employment and periodic medicals I see results shaped by the job itself: raised haematocrit in men working long hours in heat with poor fluid access, raised creatinine after a shift of heavy manual work in summer that normalizes completely once rehydrated, and raised liver enzymes in workers with solvent exposure and no protective equipment.
Shift workers show consistently worse lipid and glucose profiles than day workers doing the same job, and that is a real effect of circadian disruption rather than a character flaw. Where lead, cadmium, solvents or noise exposure exists, biological monitoring is a legal duty in many countries under ILO and OSHA-equivalent frameworks, and it is separate from your personal health screening.
Questions worth asking your own doctor#
- Which of these results actually changes what we do?
- Does anything here need repeating, and when?
- Is there a pattern across my last few reports, or is this a one-off?
- Are any of my medicines or supplements affecting these numbers?
- What symptom or change should make me come back sooner than planned?
The bottom line#
Reference ranges describe populations, not individuals, so a value just outside one is common and usually unremarkable. Read your results as a pattern across tests and across time, not as a list of pass-fail lines. The handful that most often change decisions are hemoglobin with ferritin, HbA1c, kidney function and thyroid function. If something is genuinely wrong, it rarely hides in one marginal number. It shows up as a coherent story that matches how you feel.
Common questions
My result is flagged red but only just outside the range. Should I worry?
Why do the normal ranges differ between two laboratories?
Do I need to fast before a blood test?
What is the difference between HbA1c percent and mmol/mol?
My hemoglobin is normal, so my iron must be fine?
Should I pay for a large private health panel?
How long do results take to change if I change my lifestyle?
Sources
Still not sure what this means for you?
Bring your reports to a call with one of our doctors. Leave with a written summary and the right questions for your own doctor.
