Biomarkers

TSH Blood Test: Why It Rises When Your Thyroid Slows Down

Medically reviewed by Longevity Benchmark Editorial Last reviewed 2026-08-16

A TSH blood test measures a pituitary hormone, not a thyroid one. Labs call 0.45–4.50 mIU/L normal; an optimal target sits nearer 1.0–2.0. Free T4 completes it.

TSH, thyroid stimulating hormone, is made by the pituitary gland at the base of the brain rather than by the thyroid. That one fact reorganises everything else about a TSH blood test. The number rises when thyroid output falls and falls when thyroid output rises, because the pituitary is compensating in the opposite direction to the gland it controls. It is the most frequently ordered thyroid test in medicine and the one most often read backwards.

The Verdict

A typical lab reference range runs roughly 0.45–4.50 mIU/L, while an optimal target sits nearer 1.0–2.0 mIU/L. TSH moves inversely to thyroid output, and the relationship is log-linear: a small fall in free T4 produces a disproportionately large rise in TSH. That amplification makes TSH the sensitive first-line screen, and it is also why TSH alone is incomplete. TSH measures the pituitary's response. Free T4 measures the hormone you actually have.

What a TSH blood test physically measures

A TSH blood test measures the concentration of thyroid stimulating hormone in serum, reported in milli-international units per litre. The anterior pituitary secretes it in response to a signal from the hypothalamus, and its only job is to tell the thyroid to make and release more hormone.

The thyroid answers mostly in thyroxine, T4, which circulates almost entirely bound to carrier proteins. The small unbound fraction, free T4, is the part tissues can use, and it feeds back on the pituitary to shut TSH secretion down as it rises. Most laboratories put free T4 somewhere around 0.8–1.8 ng/dL, roughly 10–23 pmol/L, with meaningful variation between assays. Reports label it either way round, as free T4 or as "T4, free", and a t4 free blood test beats total T4 because total T4 also moves with binding-protein levels that change in pregnancy and with oestrogen.

Why TSH goes up when the thyroid slows down

TSH goes up when the thyroid slows down because the pituitary is pushing harder on a gland that is not responding. A raised TSH does not mean the pituitary is broken. It usually means the opposite: the feedback loop is working as designed and reporting a shortfall downstream.

The part most pages omit is that the pituitary does not respond to free T4 in a straight line. The relationship is log-linear, so each small drop in free T4 produces a much larger proportional rise in TSH. Across the full pathological range, roughly a twofold change in free T4 corresponds to something on the order of a hundredfold change in TSH. That amplifier is what makes TSH the sensitive screen: free T4 can still be inside its reference range while TSH has already moved several-fold from where it used to be. It also means the population band is far wider than any one person's variation, since everyone has a narrow individual set point. A TSH climbing from 1.1 to 3.9 across three annual panels never leaves the printed range, never triggers a flag, and describes a real change in how hard the pituitary is working. Your own previous result is the better comparator, an argument set out in normal vs optimal ranges.

TSH levels: standard ranges and what each band suggests

Band	TSH	What it suggests
Low or suppressed	Below roughly 0.45 mIU/L	The pituitary has throttled back, usually because thyroid hormone is high. Illness, early pregnancy and assay interference also produce it.
Typical lab reference range	Roughly 0.45–4.50 mIU/L	The band nearly every laboratory prints as normal. Exact limits vary by assay, and there is no single universal range.
Optimal target	About 1.0–2.0 mIU/L, argued as wide as 0.5–2.5	Where thyroid-antibody-negative people cluster. An argued target rather than a diagnostic cut-off, and a contested one.
Upper-normal grey zone	About 2.5–4.50 mIU/L	Inside "normal" at almost every lab. A TSH of 4.2 sits here, and it is where many people report fatigue and cold intolerance.
Above range	Above roughly 4.50 mIU/L	Free T4 decides whether this is overt or subclinical hypothyroidism. Values above about 10 mIU/L warrant assessment rather than a wait-and-see repeat.

Low TSH: what suppresses the signal

Cause	How often it explains a result	Notes
Hyperthyroidism	The most common reason by far	Graves' disease and toxic nodules push thyroid hormone up, and the pituitary responds by shutting off TSH almost completely.
Thyroiditis	Common and self-limiting	Inflammation releases stored hormone in a burst, suppressing TSH for weeks. The same person may swing to a high TSH months later.
Acute or severe illness	Common in hospital, easy to miss otherwise	Infection, surgery and starvation suppress TSH transiently, and it can rebound above range during recovery.
Early pregnancy	Expected, not abnormal	hCG weakly stimulates the TSH receptor, so first-trimester TSH is normally lower. Pregnancy uses its own reference ranges.
Pituitary or hypothalamic disease	Uncommon, and the one not to miss	Here TSH is low or unremarkably normal while free T4 is low, because the signal itself has failed rather than the gland.

The distinction that matters most here is between a suppressed TSH caused by too much thyroid hormone and a low TSH caused by a pituitary that has stopped signalling. On a one-line report they look the same and mean opposite things. Free T4 separates them immediately: high with the first, low with the second.

High TSH: what makes the pituitary push harder

Cause	How often it explains a result	Notes
Autoimmune thyroiditis (Hashimoto's disease)	The dominant cause in iodine-sufficient countries	Thyroid peroxidase (TPO) antibodies identify it, and their presence changes how closely a borderline result is followed.
Subclinical hypothyroidism	Common, and more frequent with age	TSH above range with a normal free T4. Genuinely contested territory rather than an automatic diagnosis.
Recovery from acute illness	Situational and frequently misread	TSH commonly overshoots above the reference range for a few weeks after a significant illness, then settles on its own.
Age	Very common and under-appreciated	The population distribution of TSH drifts upward with age. A mildly raised value at 85 does not mean the same thing it means at 35.
Medications	Uncommon but readily checked	Lithium, amiodarone and checkpoint-inhibitor immunotherapy are recognised causes. So is inadequate absorption of a prescribed thyroid medicine.
Assay artefact	Rare, and worth knowing exists	Heterophile antibodies and macro-TSH can produce a persistently raised TSH in someone with no thyroid disease at all. The lab can test for it.

TSH and free T4 read together: the patterns that matter

TSH and free T4 read together sort almost every thyroid result into one of five patterns, and the pattern rather than either number alone determines the next step.

Pattern	TSH	Free T4	What it usually means
Overt hypothyroidism	High	Low	The thyroid cannot meet demand despite maximum pituitary signalling. The clearest indication for treatment, and a clinician decision.
Subclinical hypothyroidism	High	Normal	Output is being maintained, but only because the pituitary is pushing harder. Antibodies, symptoms, age and pregnancy status decide what happens next.
Overt hyperthyroidism	Low or suppressed	High	Too much thyroid hormone, so the pituitary switches the signal off. Needs assessment, particularly with palpitations or weight loss.
Subclinical hyperthyroidism	Low	Normal	Often transient. Persistent suppression carries atrial fibrillation and bone-density concerns, so it is followed rather than ignored.
Central hypothyroidism	Low or unremarkably normal	Low	The uncommon pattern that points at the pituitary or hypothalamus rather than the thyroid. TSH screening alone can miss it entirely.

The last row is why a free T4 blood test earns its place on the requisition. Screening on TSH alone works because the pituitary is almost always the healthy part of the loop. When it is not, a low free T4 sits beneath a TSH that looks unremarkable, and a protocol stopping at TSH reports a normal thyroid function test to someone with genuine hypothyroidism. It is uncommon, and it is the one failure mode TSH-only screening cannot detect by design.

The optimal band, and why subclinical hypothyroidism is contested

An optimal TSH band is an argument about where symptoms and risk cluster, not a diagnostic threshold. A reference range is built by measuring a population and printing the central 95%, and that method asks whether a value is common, never whether it is good. Older thyroid reference populations included people with undiagnosed autoimmune disease. Screen those individuals out and the upper limit in disease-free groups falls closer to 2.5–3.0 mIU/L, with the cluster where people report feeling well nearer 1.0–2.0. Where an optimal ceiling should actually sit is unsettled, and arguments for a wider 0.5–2.5 band are made on the same evidence, which is why we treat the target as a direction rather than a line. A TSH of 4.2 is inside the printed range at essentially every lab, and it is also squarely in the band where fatigue and cold intolerance are commonly reported.

None of that settles what to do about it, and this page will not pretend otherwise. Subclinical hypothyroidism, a raised TSH with a normal free T4, is genuinely unresolved. Four things change the answer: whether TPO antibodies are present, since autoimmune disease progresses to overt hypothyroidism at a higher yearly rate; whether there are symptoms attributable to thyroid function; age, because the TSH distribution shifts upward in later life and a mildly raised value at eighty may carry no benefit from intervention; and pregnancy, where demand rises and separate reference ranges apply. Weighing those four is a clinical decision, and this page gives no guidance on thyroid medication or dosing.

Three things that move a TSH blood test result without the thyroid changing

Three ordinary circumstances shift a TSH blood test result enough to reclassify a borderline value, and none involves any change in the thyroid.

Draw time. TSH has a diurnal rhythm. It peaks overnight and in the early hours and falls to a trough in the late afternoon and early evening. A borderline value drawn at 8 a.m. and repeated at 4 p.m. can land on opposite sides of the reference limit with nothing else different, which is why a "changed" TSH is so often a changed appointment slot.
Acute illness. Infection, surgery and other significant physiological stress suppress TSH while the illness runs, and TSH then commonly overshoots above the reference range during recovery. A panel drawn in either window describes the illness, not the gland. Waiting several weeks after recovery gives an interpretable result.
Biotin. High-dose biotin interferes with the streptavidin-biotin chemistry many immunoassays are built on, and it distorts thyroid results in either direction depending on the assay format. Sandwich assays such as TSH read falsely low; competitive assays such as free T4 read falsely high. The classic outcome is a laboratory picture of hyperthyroidism in someone with a normal thyroid. This is analytical interference, not a biological effect, which is why laboratories commonly ask people to pause high-dose biotin for around 48 to 72 hours before a thyroid panel. Tell whoever ordered the test what you take, hair and nail supplements included, rather than changing anything yourself.

A rarer fourth explanation exists. Heterophile antibodies and macro-TSH, an inactive complex of TSH bound to immunoglobulin, can produce a persistently raised TSH in someone with entirely normal thyroid function. When a high TSH keeps repeating with a normal free T4, no antibodies and no symptoms, asking the laboratory to check for interference is cheaper than another year of follow-up.

What TSH means in combination

High TSH + low free T4: overt hypothyroidism. This is the pattern with the least ambiguity, and the one that most clearly warrants a clinician conversation rather than watchful waiting.
High TSH + normal free T4 + positive TPO antibodies: autoimmune thyroid disease with output still maintained. The antibody result is what changes the follow-up interval, since progression is more likely.
High TSH + raised MCV: hypothyroidism enlarges red cells, and the macrocytosis usually resolves once thyroid status is corrected. An unexplained high MCV is a reason to check thyroid function before assuming B12.
High TSH + rising ApoB or LDL cholesterol: an underactive thyroid raises atherogenic particle counts. Treating the lipid result alone misses a correctable driver.
Low TSH + high free T4 + high SHBG: thyroid hormone drives SHBG upward, so a rising SHBG with no other explanation supports thyroid excess.
High TSH + mildly raised ALT: thyroid disease is a recognised and readily testable cause of a modestly abnormal liver enzyme.
Low or unremarkable TSH + low free T4: the central pattern. Investigation moves to the pituitary, and the other pituitary hormones become the relevant tests.

How and when to take a TSH blood test

Draw at a consistent time in the morning. The diurnal rhythm is large enough that comparing a morning result against an afternoon one produces trends that are not real. TSH also reads slightly higher fasting than after a meal.
Order free T4 alongside TSH, from the start if there are symptoms or as a reflex if TSH falls outside the range. Without it, the overt-versus-subclinical question and the central pattern go unanswered.
Add TPO antibodies once if TSH is above range, since autoimmune disease is the usual cause and antibody status changes how a borderline result is followed. Repeating them annually adds little.
Do not test during or just after an acute illness, which distorts thyroid results in either direction depending on which phase you catch.
List your supplements and medications for whoever ordered the test, particularly high-dose biotin, lithium and amiodarone.
Repeat before concluding anything, allowing six to eight weeks between draws, and use the same laboratory so assay differences do not masquerade as change. Our biomarker guides cover how the standard panel groups fit together, the practical way to get free T4 and antibodies onto the same requisition.

When a result warrants seeing a physician

A TSH above roughly 10 mIU/L, which is handled differently from a mildly raised value and should not wait for a routine annual repeat.
Any raised TSH with a low free T4, the overt hypothyroid pattern.
A low or unremarkably normal TSH with a low free T4, particularly with headache, visual field changes or other pituitary hormone abnormalities.
A suppressed TSH with palpitations, an irregular pulse, tremor, heat intolerance or unintended weight loss.
Any abnormal thyroid result in pregnancy or when planning one, where different reference ranges apply.
A neck lump, a visibly enlarged thyroid, or new difficulty swallowing, whatever the TSH says.
Before starting any prescription therapy on the strength of an optimal-range argument, which is a decision that needs a clinician rather than a target number.

A TSH blood test is the right first question and a poor last word. It is sensitive, cheap and correctly placed at the front of every thyroid workup, and it reports the pituitary's response rather than your thyroid hormone level. Read it with free T4, read the pair as a pattern, and leave the interpretation with a clinician who can see the antibodies, the symptoms, your age and the rest of the panel.

Frequently Asked Questions

What is TSH in a blood test?

TSH stands for thyroid stimulating hormone, and it is made by the pituitary gland, not by the thyroid. Its job is to instruct the thyroid to produce more hormone, so the level reads as an inverse measure of thyroid output: TSH goes up when the thyroid is underperforming and down when it is overproducing. A TSH blood test therefore tells you how hard the pituitary is having to push. It does not directly measure how much thyroid hormone is in your blood, which is what free T4 measures.

What is a normal TSH level?

Most laboratories report roughly 0.45 to 4.50 mIU/L as normal for non-pregnant adults, with real variation between assays and reference populations. An optimal target nearer 1.0 to 2.0 mIU/L is commonly argued for, on the basis that the wider band was derived from populations that included people with undiagnosed autoimmune thyroid disease. That argument is about where symptoms and risk cluster, not about a diagnostic threshold, and whether the lower ceiling should be used clinically is still disputed among endocrinologists.

What does a high TSH mean?

A high TSH means the pituitary is working harder than usual to get thyroid hormone produced, which points towards an underactive thyroid. Free T4 splits the result in two: a high TSH with a low free T4 is overt hypothyroidism, and a high TSH with a normal free T4 is subclinical hypothyroidism. Autoimmune thyroiditis is the usual underlying cause, and TPO antibodies identify it. A single raised TSH is also commonly a passing artefact of recent illness, so it is normally confirmed on a second draw before anything is concluded.

What does a low TSH mean?

A low TSH usually means there is too much thyroid hormone in circulation, so the pituitary has cut the signal back. The common causes are hyperthyroidism, a burst of stored hormone released by thyroiditis, and thyroid hormone taken as medication. Acute illness and the first trimester of pregnancy both lower TSH without any thyroid problem. The exception that matters is a low TSH accompanied by a low free T4, which points at the pituitary rather than the thyroid and needs a different investigation entirely.

What is a free T4 blood test and why is it ordered with TSH?

A free T4 blood test measures the unbound fraction of thyroxine circulating in your blood, which is the portion available to tissues. It answers the question TSH cannot: how much thyroid hormone you actually have. Total T4 is reported by some panels but moves with binding-protein levels, which change in pregnancy and with oestrogen, so free T4 is the more reliable of the two. TSH and free T4 are read as a pair because the combination, not either number alone, separates overt from subclinical disease and thyroid problems from pituitary ones.

Does the time of day affect a TSH blood test?

Yes, and enough to move a borderline result. TSH follows a daily rhythm, peaking overnight and in the early hours and falling to its lowest point in the late afternoon and early evening. A morning draw and an afternoon draw on the same person can straddle the top of the reference range, which means a value flagged on one occasion and cleared on another may reflect the clock rather than the thyroid. Drawing at a consistent time, and comparing like with like, removes a large share of apparent change between panels.

Can biotin supplements affect thyroid blood tests?

Yes. High-dose biotin, the kind found in hair, skin and nail products and in some high-strength B-complex supplements, interferes with the streptavidin-biotin chemistry used by many immunoassays. The direction depends on the assay format: sandwich assays such as TSH tend to read falsely low, while competitive assays such as free T4 tend to read falsely high, so the classic artefact is a laboratory picture of hyperthyroidism in someone whose thyroid is fine. This is assay interference, not a biological effect. Laboratories commonly ask people to pause high-dose biotin for around 48 to 72 hours before a thyroid panel, and the right move is to tell whoever ordered the test what you take rather than to change anything on your own.