The rest of this page stays on the labs, the patterns that separate the two directions, the patterns that blur them, and what a clinician may add when the panel does not read cleanly.
TSH is the pituitary's instruction to the thyroid, not a measure of thyroid hormone itself. The pituitary reads how much thyroid hormone is circulating, then raises or lowers its TSH output to push the gland harder or ease off. Because that adjustment is continuous and sensitive, TSH often moves outside its range while free T4 still reads normal, which is why thyroid panels usually lead with TSH and read everything else against it 1. A single value is a snapshot of signalling rather than a verdict on the gland. Full detail on the marker, how the sample is run, its reference ranges and how to read one result in context sits on the TSH guide. Everything below is about how the two directions differ from each other.
TSH is a corrective signal, so it rises when thyroid hormone is scarce and falls when thyroid hormone is plentiful. The hypothalamus releases TRH, the pituitary responds by releasing TSH, the thyroid responds by releasing mostly T4 along with a smaller amount of T3, and tissues convert T4 into the more active T3. Circulating hormone then feeds back on the pituitary and damps the signal. That is the loop, and it is the reason a lab report reads as a mirror.
The mirror is amplified rather than proportional. Research suggests that the relationship between TSH and free T4 is log-linear, so a modest fall in free T4 produces a proportionally much larger rise in TSH, which makes TSH the earliest moving number in most thyroid panels 1. The same amplification works in reverse, which is why TSH can be fully suppressed while free T4 is only slightly raised.
Two conditions have to hold for the mirror to be readable. The pituitary has to be working normally, and the axis has to be at steady state. TSH re-equilibrates slowly after thyroid hormone levels change, so a result drawn soon after a shift can describe where the axis was rather than where it is 1. Both exceptions get their own sections below, because they account for most of the panels that look contradictory at first glance.
A raised TSH with a free T4 at or below the bottom of its reference range is the pattern that points toward an underactive thyroid. Research suggests that primary hypothyroidism, meaning a problem in the gland itself rather than in the pituitary, is identified biochemically by an elevated TSH together with a low free T4, and that it accounts for the large majority of underactive presentations 2.
The sequence matters more than any single number. Because the pituitary reacts first, TSH typically climbs while free T4 is still inside its range, producing the milder pattern described later on this page. Free T3 is often the last value to fall, because the body preferentially converts and conserves the active hormone, so a normal free T3 does not rule the pattern out.
One result is not a conclusion. Research suggests that an abnormal TSH should be confirmed with repeat testing that includes free T4 before any interpretation is settled, because transient shifts are common and a single draw can mislead 3. Symptoms overlap heavily between the two directions and between thyroid and non-thyroid causes, which is precisely why the panel, not the symptom list, does the separating work. What a raised TSH with a low free T4 means for you personally, and what should happen next, is a conversation for a clinician who can see the whole picture.
A low or fully suppressed TSH with a free T4 or free T3 above range is the pattern that points toward an overactive thyroid. Studies indicate that thyrotoxicosis is defined biochemically by a suppressed TSH alongside raised free thyroid hormone levels, with the suppression reflecting a pituitary that has stopped signalling because circulating hormone is already excessive 4.
The degree of TSH suppression carries information. A TSH that is low but still detectable sits differently from one that is undetectable, and the frees decide whether the pattern is overt or mild. Guidance regards the measurement of free T4 together with total or free T3 as part of the initial biochemical assessment once TSH is suppressed, because the two hormones do not always rise together 5.
The panel establishes the direction, not the reason. A suppressed TSH with raised frees can arise from a gland that is overproducing, from inflammation releasing stored hormone, or from hormone arriving from outside the gland, and distinguishing those requires further testing that a clinician selects, including antibody measurement and, where appropriate, uptake imaging 5. None of that is decided from a printout. Any next step, including whether any treatment is appropriate, is a decision for a licensed provider after a full evaluation.
The cleanest way to hold the comparison is as a grid of three numbers, TSH, free T4 and free T3, read together rather than one at a time. The table below summarises the recognised combinations and what each may reflect. It describes patterns, not diagnoses.
| Lab pattern | TSH | Free T4 | Free T3 | What it may reflect |
|---|---|---|---|---|
| Overt underactive | High | Low | Normal or low | Primary hypothyroidism, worth confirming on a repeat panel |
| Mild (subclinical) underactive | High | Normal | Normal | Early or mild reduction in thyroid output |
| Overt overactive | Low or suppressed | High | High | Thyrotoxicosis from one of several causes |
| Mild (subclinical) overactive | Low or suppressed | Normal | Normal | Mild excess signalling, or a transient shift |
| T3 predominant overactive | Suppressed | Normal | High | An overactive pattern that free T4 alone would miss |
| Central (pituitary or hypothalamic) | Low, normal or only mildly raised | Low | Low or normal | A signalling problem above the thyroid |
| Illness and recovery | Low during illness, may overshoot high afterwards | Normal or low | Low | The effect of acute illness on the axis rather than thyroid disease |
| Discordant or implausible | Any combination that does not fit | Discordant | Discordant | Worth checking for assay interference before anything else |
Read the grid in that order. TSH gives the direction, the frees give the severity, and anything that breaks the mirror sends you to the lower half of the table.
The numeric boundary that separates the two directions sits in TSH, with the free hormones deciding how far the pattern has travelled. Standard reference ranges and FitScript's optimal range are shown below.
| Marker | Typical lab reference range | FitScript optimal range |
|---|---|---|
| TSH | 0.45 to 4.5 mIU/L | 0.5 to 2 mIU/L |
FitScript's optimal range is an optimization target, not a diagnostic cut-off, and a result outside it is not a diagnosis of anything. Typical lab reference ranges vary by laboratory, assay, age and sex, and pregnancy uses its own ranges.
Free T4 and free T3 are deliberately left without numbers here. Their reference ranges differ meaningfully between assays and laboratories, so the only ranges worth reading them against are the ones printed beside the result on your own report. Comparing a free T4 from one laboratory with a range from another is one of the commonest ways a normal panel gets misread as suboptimal, and the reverse.
Two practical consequences follow. First, direction and trend across repeat draws at the same laboratory tell you more than a single number compared with a published figure. Second, a value just outside a range and a value far outside it are different conversations, and neither is settled from one draw.
A subclinical pattern is one where TSH is outside its reference range while free T4 and free T3 are both still inside theirs. It is the most common reason a panel looks like it belongs to neither category cleanly.
On the underactive side, a raised TSH with a normal free T4 is the mild pattern, and research suggests that confirming it with repeat testing before drawing conclusions is important because TSH fluctuates and a proportion of raised values normalise on retesting 3. On the overactive side, a low or suppressed TSH with normal free T4 and free T3 is associated with subclinical hyperthyroidism, and guidance regards the degree of suppression and whether it persists on a repeat draw as the key questions 5.
The reason these patterns blur the line is that they are genuinely intermediate. The gland is producing enough hormone to keep the frees in range, and the pituitary is working harder or backing off to make that happen. Whether that state deserves monitoring, further testing or nothing at all depends on the size of the shift, whether it persists, antibody status, age, pregnancy status and the rest of the clinical picture. That judgement belongs to a clinician, and this is one of the clearest cases where a single result on its own cannot carry a decision.
When TSH and free T4 move together rather than opposite each other, the problem is usually not in the thyroid. A low or inappropriately normal TSH alongside a low free T4 points above the gland, to the pituitary or hypothalamus, and the pattern is easy to miss because a TSH-first strategy reads the TSH as acceptable 7. This is one of the few situations where measuring free T4 regardless of the TSH value changes what the panel shows.
The opposite concordant pattern, a normal or raised TSH with a raised free T4, is rare and has several possible explanations, including laboratory interference, a pituitary cause and inherited differences in how tissues respond to thyroid hormone 7. Because the possibilities range from a purely technical artefact to something that needs specialist assessment, the sensible first move is not to interpret it at all, but to confirm it.
In practice, a concordant panel should prompt three questions. Was the sample drawn at steady state, or shortly after something changed. Could the assay be affected by interference, which the next sections cover. And does the clinical context point above the thyroid rather than at it. A clinician decides which of those to pursue and in what order, and a discordant panel is worth raising promptly rather than filing away.
Acute illness changes thyroid numbers without the thyroid being the problem, which is why timing a test badly can produce a pattern that imitates either direction. Free T3 usually falls first during significant illness, free T4 may follow, and TSH can drop into the low range while the body is unwell and then overshoot above its range during recovery 7.
That creates two traps. A panel drawn during illness, with a low TSH and a low free T3, can be mistaken for an overactive pattern even though the frees point the other way. A panel drawn in the weeks after recovery, with a transiently raised TSH, can be mistaken for an underactive pattern. Reading TSH in isolation makes both errors more likely, because the frees are what reveal the inconsistency 1.
The practical rule is simple. Where it can be avoided, thyroid panels are best drawn when you are well and at steady state, and an abnormal panel drawn around an illness, a hospital stay, a surgical recovery or a major physiological stress deserves a repeat once things settle rather than an immediate conclusion. If the result is already in hand, note the circumstances of the draw when you discuss it, because that context changes how the numbers should be read.
Several technical and physiological factors move thyroid numbers without any change in gland function, and recognising them prevents a panel being over-interpreted. High-dose biotin supplementation can interfere with the immunoassays used for thyroid testing and produce results that imitate an overactive pattern, with a falsely low TSH and falsely raised free hormone readings. Antibodies in the sample that cross-react with assay components can distort results in either direction 7.
Binding proteins matter too. Total T4 and total T3 move with changes in thyroid binding globulin, which shifts in pregnancy and with oestrogen status, whereas the free measurements are designed to be less affected, so a panel reporting totals can look abnormal when free hormone is unchanged 7. TSH also varies through the day and is typically higher overnight and in the early morning than in the afternoon, so draw time contributes some of the variation seen between repeat tests 1.
Four habits reduce the noise. Use the same laboratory for repeat testing where possible. Draw at a consistent time of day. Tell whoever orders the test about supplements you take, including high-dose biotin, and ask whether any should be paused before the draw and for how long, which is a clinician's call rather than a general rule. And treat any result that looks biochemically implausible as a reason to repeat before acting.
Free T3 adds most in the overactive direction and least in the underactive one. Some overactive presentations raise T3 before or without raising T4, so a panel showing a suppressed TSH and a normal free T4 can still show a raised free T3, and guidance regards measuring T3 as part of the assessment of a suppressed TSH for exactly this reason 5. Omitting it is one way a mild looking overactive panel gets filed as subclinical when it is not.
In the underactive direction, free T3 is less informative as a screening value.
Free T3 is also the value most disturbed by non-thyroid factors, since it falls during acute illness independently of thyroid function 7. That combination, highly informative in one specific scenario and easily confounded in others, is why free T3 belongs on a panel as a third reading rather than as a standalone marker. It is read alongside TSH and free T4, never instead of them.
Antibody testing addresses cause rather than direction, and which antibodies are useful depends on which way the panel points. In an underactive pattern, thyroid peroxidase antibodies and, less often, thyroglobulin antibodies are the markers used to identify autoimmune thyroiditis as the underlying mechanism 2. A positive result does not change the TSH or free T4 reading, but it does change how a borderline pattern is understood and how closely it is followed.
In an overactive pattern, TSH receptor antibodies are the test that distinguishes an autoimmune cause from the other reasons a TSH can be suppressed, and current guidance regards antibody measurement, with uptake imaging where antibody testing is unavailable or inconclusive, as the route to establishing the cause 6. Establishing cause matters because inflammation that releases stored hormone and a gland that is overproducing can look similar on the first panel yet behave very differently over time 5.
Two limits are worth knowing. Antibody levels do not grade how underactive or overactive a thyroid is, so they are not a severity score. And antibody results are interpreted alongside the hormone panel and the clinical picture, not in place of it. A clinician decides which antibodies to order and when, and the result is one more input into that assessment.
All three core values come from one ordinary blood draw, with antibodies added to the same sample when a clinician orders them. Fasting is not required for thyroid hormones, though many people draw thyroid markers alongside fasting markers in the same appointment, which is fine as long as the timing stays consistent between tests. Because TSH is higher overnight and early in the morning than later in the day, keeping draw time and laboratory consistent makes repeat results comparable.
Repeat timing is a clinical decision rather than a fixed rule, and it depends on how far the result sits from range, whether symptoms are present and what else is going on. What is consistent across guidance is that an abnormal TSH is confirmed on a repeat panel that includes free T4 before conclusions are drawn, because single values fluctuate 3. Because TSH re-equilibrates slowly, repeating too soon after any change in the axis can show a number that is still catching up 1.
FitScript members have thyroid markers run as part of 150+ biomarkers tested per year, and Atlas, FitScript's AI health advisor, reads them against the rest of the panel rather than as an isolated three-line result. Interpretation in context is the point. A TSH that sits slightly above optimal alongside stable free hormones and no antibodies is a different observation from the same TSH with a falling free T4.
Treat an in-between result as an observation that needs a second data point, not as a verdict. The structure that keeps this honest is simple. What does the value show, what might it reflect, and what is the next step worth raising with a clinician.
What it shows is a number relative to a range on one day, at one laboratory, from one draw. What it may reflect is a long list, including an early shift in thyroid output, recovery from an illness, assay interference, normal biological variation, draw timing, or an issue above the gland. What it is not is a diagnosis, because no single value identifies a condition and the mild patterns in particular resolve on repeat testing often enough that acting on one draw is a mistake 3.
The next step is almost always the same. Keep the full panel together, including free T4 and free T3 if they were run, note when the draw happened, what supplements you were taking and whether you had been unwell recently, and bring all of that to a clinician rather than the single flagged number. Do not ignore an out-of-range result, and do not self-diagnose from it either. Both errors come from reading one line instead of the pattern.
Any result outside the standard reference range is worth discussing with a clinician, and some are worth raising promptly. A markedly suppressed TSH with free T4 or free T3 above range, a clearly raised TSH with a low free T4, a panel where TSH and free T4 move in the same direction, or any abnormal thyroid result in pregnancy or while planning a pregnancy all fall into the category of worth discussing with a clinician promptly rather than at the next routine visit.
It is also worth a conversation when the numbers and how you feel do not agree in either direction, when results swing noticeably between draws, or when a previously stable pattern changes. Symptom overlap is wide enough that the panel and the clinical picture need to be read together by someone who can see both.
This page is educational and not medical advice. Discuss your own results, and any treatment decision that follows from them, with a clinician or licensed provider who can evaluate whether a given option is appropriate for you.
If you want to go deeper on either direction, the guide to hypothyroidism symptoms covers how an underactive pattern can present, and does hypothyroidism affect libido looks at one specific question that comes up often once an underactive pattern is on the table.
The direction of TSH. An underactive thyroid pattern usually shows a high TSH, because the pituitary is pushing the gland harder. An overactive pattern usually shows a low or fully suppressed TSH, because the pituitary has backed off. Free T4 and free T3 then show how far the pattern has gone.
Usually it gives the direction, but not the full picture. TSH cannot tell you how far the shift has travelled, whether the cause sits in the thyroid or in the pituitary, or whether illness or assay interference moved the number. Free T4, and free T3 when TSH is suppressed, are what complete the pattern.
That combination is the mild or subclinical overactive pattern. It means the pituitary is signalling less while circulating hormone is still inside its range. It is not a diagnosis. The usual next step is a repeat panel after an interval, and a clinician decides whether anything further is needed.
Fasting is not required for thyroid hormones. Timing does matter, because TSH is higher overnight and early in the morning than later in the day. Drawing at a similar time of day, at the same laboratory, makes repeat results easier to compare.
That is a clinical decision, not a fixed interval. It depends on how far the value sits from range, whether you were unwell around the draw and what the free hormones showed. Ask whoever ordered the test when to repeat it, and keep the same laboratory for the recheck.
Yes, a panel can look different on a later draw, which is one reason single results are not treated as final. Illness, recovery, natural variation and changes in the gland itself can all move the numbers. Serial panels interpreted by a clinician show far more than one snapshot.
High-dose biotin can interfere with the assays used for thyroid testing and produce results that look like an overactive pattern when the thyroid is unchanged. Tell whoever orders the test what you take, and ask whether anything should be paused before the draw and for how long.
Antibodies help identify the cause behind a pattern rather than its severity. Different antibodies are useful depending on whether the panel points toward an underactive or an overactive thyroid. They are ordered and interpreted by a clinician alongside the hormone results, never read on their own.
A licensed provider does, after reviewing your full panel, your history and anything else relevant. Testing describes what your biology is doing. It does not decide whether any intervention is appropriate, and no result on this page should be read as a diagnosis or a recommendation.