Can a TSH Test Be Wrong? What Influences Your Results

A TSH test can absolutely produce a result that does not reflect your actual thyroid status. The number on your lab report is influenced by what time you had blood drawn, whether you ate beforehand, your age, supplements you take, and even the specific machine your lab used to run the test. In one study, eating before a morning blood draw dropped TSH by roughly a third compared to a fasting sample, enough to reclassify people from mildly hypothyroid to normal. That kind of swing means a single TSH value, taken out of context, can genuinely mislead both you and your doctor.

The Time You Get Your Blood Drawn Changes the Number

TSH follows a circadian rhythm, peaking in the late evening and early morning hours and falling to its lowest point in the afternoon. This is not a subtle fluctuation. Research on circadian and seasonal rhythms in thyroid hormones shows that the upper limit of the TSH reference range shifts meaningfully depending on when the sample is collected, and that the nighttime TSH surge grows more pronounced with age.¹ A separate study looking specifically at the practical impact of draw time concluded that standardizing when blood is collected is necessary for TSH measurements to be comparable across visits.²

What this means for you: if your first TSH was drawn at 7 a.m. and the follow-up was drawn at 2 p.m., comparing the two numbers side by side is misleading. TSH naturally runs lower later in the day, so an afternoon draw may look like improvement (or like your thyroid is fine) when really the clock just shifted the reading. If you are tracking TSH over time, try to get blood drawn at roughly the same hour each visit.

Eating Before the Test Suppresses TSH

Food intake pushes TSH down, and the effect is large enough to change a diagnosis. A controlled study found that people who ate before a 9 a.m. blood draw had a median TSH drop of about 32% compared to their 7 a.m. fasting level. Even a glucose-only drink produced a similar drop. In a fasting group, TSH still fell slightly between 7 and 9 a.m. (about 5%), consistent with the normal circadian decline, but the food-driven drop was six times bigger. Out of nine subjects who met criteria for subclinical hypothyroidism based on their fasting morning sample, only two still qualified after eating.³

A separate study found that 75% of subjects who looked mildly hypothyroid on a fasting draw were reclassified as normal when tested after a meal.⁴ That same study recommended fasting samples as the standard for TSH, since a normal fasting result removes the need for repeat testing. An earlier study confirmed that TSH values obtained during extended fasting were significantly lower than standard fasting values, further complicating interpretation.⁵ The practical advice here is straightforward: get your TSH drawn fasting, first thing in the morning. This gives you the most reliable, reproducible number.

Age Shifts What Counts as “Normal”

Most labs print a single TSH reference range on your report, usually something like 0.4 to 4.5 mIU/L, and flag anything outside it. But this one-size-fits-all range does not account for the fact that TSH naturally drifts upward as people get older. In a large study of elderly adults, age-specific upper limits for TSH were 5.51 mIU/L in the 65-to-70 age group and rose to 6.70 mIU/L in people over 80. When researchers applied these age-adjusted ranges instead of the standard lab cutoff, the apparent prevalence of subclinical hypothyroidism dropped from about 10% to under 4%.⁶

Another study of elderly populations reported similar findings: the upper TSH limit was about 5.1 mIU/L for older men and 5.25 mIU/L for older women, both well above the general-population cutoff of 4.31 mIU/L. When the age-specific range was applied, the prevalence of subclinical hypothyroidism fell by roughly 3.5 percentage points.⁷ The implication is that many older adults are being told their thyroid is underactive when their TSH is simply doing what TSH does in their age group. Experts have argued that starting lifelong levothyroxine treatment based on a single mildly elevated TSH in an older person, without repeat testing or age-adjusted interpretation, can be unnecessary and even harmful.⁸

Pregnancy Requires Its Own Reference Ranges

During pregnancy, the body’s demand for thyroid hormones rises, and the hormone hCG (which surges in the first trimester) stimulates the thyroid directly. TSH tends to drop during early pregnancy, sometimes to levels that would look abnormally low in a non-pregnant person. Both free thyroid hormone and TSH reference intervals shift throughout pregnancy, and most professional guidelines now recommend using method-specific and trimester-specific reference ranges to interpret results in pregnant patients.⁹ If your doctor is comparing your first-trimester TSH to a standard non-pregnant range, the result could easily be flagged as abnormal when it is a perfectly normal physiological shift.

Biotin Supplements Can Fake Thyroid Disease

This one catches people off guard. Biotin (vitamin B7), found in hair-skin-nails supplements and prescribed at high doses for conditions like multiple sclerosis, interferes directly with the chemistry that many thyroid immunoassays rely on. In a published case report, a patient taking high-dose biotin for MS had lab results that looked like textbook thyrotoxicosis: suppressed TSH, elevated free T4. The patient was not hyperthyroid. After stopping biotin for one week, both TSH and free T4 returned to normal.¹⁰

The interference stems from the biotin-streptavidin binding system used in many commercial immunoassays. Not every assay is equally vulnerable. One study using a Beckman analyzer found that TSH and total T4 assays that did not use biotin-based chemistry were unaffected, while biotin-dependent assays on the same platform showed significant interference.¹¹ Even over-the-counter doses of biotin (5,000 to 10,000 mcg, commonly sold for hair growth) have been reported to cause interference, though the most dramatic cases involve therapeutic doses for neurological conditions. If you take any biotin-containing supplement, mention it to your doctor before thyroid testing. Most labs recommend stopping biotin at least two to three days before a blood draw.

Antibodies That Fool the Assay

Your blood can contain antibodies that directly interfere with the laboratory machinery measuring TSH. These are called heterophilic antibodies, and they are not related to thyroid disease. They bind to the reagents in the immunoassay and produce readings that do not reflect actual TSH levels. The interference can push TSH readings either falsely high or falsely low, though falsely elevated results are more commonly reported. Two-site immunoassays (the kind typically used for TSH) are more susceptible to this than assays for free T4 or free T3.¹² Researchers have confirmed that heterophilic antibodies can interfere with TSH measurements across multiple major analyzer platforms (Siemens, Abbott, Roche), and that the interference becomes apparent when samples are pre-treated with a heterophilic antibody blocker.¹³

A related but distinct phenomenon is macro-TSH. In some people, TSH circulates bound to an antibody (usually IgG), forming a larger complex that the body clears slowly. The TSH accumulates to very high measured levels, sometimes 30 to 50 mIU/L, yet the person has normal thyroid hormone levels, no symptoms, and a normal-looking thyroid on ultrasound.¹⁴ Macro-TSH is rare, but when present it can lead to extensive and unnecessary workups. The diagnosis is typically made using a precipitation technique or chromatography to demonstrate that the TSH in the sample is bound in a large molecular complex.¹⁵ The key red flag for clinicians is a persistently elevated TSH with completely normal free T4, free T3, and no symptoms or imaging abnormalities.

Medications That Move TSH Without Thyroid Disease

Several commonly used medications suppress TSH through direct effects on the brain’s pituitary gland or hypothalamus, without causing true thyroid dysfunction. Glucocorticoids (like prednisone), dopamine agonists (used for Parkinson’s disease and certain pituitary conditions), and somatostatin analogs all fall into this category.¹⁶ If you are on any of these and your TSH comes back low, it does not necessarily mean your thyroid is overactive. It could be the medication acting directly on the signaling pathway that controls TSH production.

For people already on thyroid medication, how and when you take levothyroxine also matters. Taking it with food rather than on an empty stomach leads to higher and more variable TSH readings, making it harder to stay in a target range.¹⁷ And inconsistent dosing creates another trap: because T4 has a long half-life, sporadic pill-taking can produce normal or even elevated thyroid hormone levels on a blood test while TSH remains stubbornly elevated, giving the impression that the dose is too low when the real problem is adherence.¹⁸

Acute Illness Distorts Thyroid Tests

When you are seriously ill with a non-thyroid condition, such as a severe infection, major surgery, or organ failure, thyroid function tests often go haywire. This pattern, sometimes called euthyroid sick syndrome, typically shows low T3, high reverse T3, and a TSH that can be low, normal, or mildly elevated, none of which reflects an underlying thyroid problem.¹⁹ Testing thyroid function during a hospitalization or acute illness is widely considered unreliable for diagnosing thyroid disease. In most cases, clinicians defer thyroid evaluation until recovery unless there is strong clinical suspicion of a thyroid emergency.

Different Labs, Different Machines, Different Numbers

Not all TSH assays are created equal. Reference intervals for thyroid function tests vary across immunoassay platforms, and this variation leads to inconsistent clinical interpretations depending on which machine your lab happens to use.²⁰ If you switch labs or if your lab changes its analyzer, your TSH might shift slightly even if nothing about your thyroid has changed. This is a recognized problem in laboratory medicine, and harmonization efforts are ongoing but incomplete. If you see an unexpected change in your TSH and you recently switched providers or labs, it is worth asking whether a different platform was used.

Seasonal Patterns and Sleep Disruption

TSH runs higher in winter and lower in summer, a pattern confirmed in a study of over 7,000 healthy people in an iodine-sufficient area.²¹ The seasonal shift is not large enough to routinely change a diagnosis, but it could nudge a borderline reading across the cutoff in either direction. A study looking at daily ambient temperature found a trend toward higher TSH in colder conditions, though the increase did not reach statistical significance.²²

Sleep disruption matters too. In a controlled experiment examining how a shifted sleep-wake cycle affects hormone rhythms, sleep deprivation amplified the nighttime TSH surge on the first night, while the surge dampened on the second day.²³ For shift workers or anyone whose sleep schedule is irregular, TSH values can be more variable than in people with a consistent sleep pattern. This variability is another reason why a single TSH measurement taken under non-ideal conditions should not be treated as definitive.

Why a Single TSH Rarely Tells the Full Story

Given everything that can push TSH around, the case for repeat testing before starting treatment is strong. A review found that about 62% of elevated TSH levels may revert to normal spontaneously, and recommended confirming subclinical hypothyroidism with a repeat test at least two months later.²⁴ A large study of older adults with mild TSH elevation found that levels normalized in about 61% of participants within a year, and even after a confirmed elevated repeat test, nearly 40% normalized over the following year without treatment. The authors suggested that a third measurement may be warranted before considering levothyroxine.²⁵ In practice, though, patients are frequently started on thyroid medication after a single elevated TSH and continue it for life, a pattern that multiple researchers have flagged as problematic.⁸

Your Genetic Set Point Is Narrower Than the Lab Range

Here is something that adds a layer of complexity most people never hear about. The standard TSH reference range is wide, typically spanning from below 0.5 to above 4.0 mIU/L, because it captures the entire population’s spread. But your individual TSH set point, the level your body naturally regulates around when everything is healthy, is much narrower than that. Heritability studies suggest that up to 67% of the variation in circulating TSH and thyroid hormone levels is genetically determined.²⁶ Your personal “normal” might be 1.2 mIU/L or it might be 3.5 mIU/L, and both could be perfectly healthy for you.

Specific gene variants influence where that set point falls. One studied polymorphism in the gene for a thyroid hormone-converting enzyme alters the sensitivity of the feedback loop between the pituitary and the thyroid, meaning people with certain variants need different TSH levels to maintain the same amount of active thyroid hormone.²⁷ This is not clinically actionable for most people yet; doctors do not routinely genotype patients to interpret their TSH. But it helps explain a common frustration: you can be well within the “normal” range and still feel off, because a TSH of 3.8 might represent your body running somewhat above its own set point, even though the lab report says everything is fine. It also means a TSH that shifts from 1.0 to 2.5 over a few years could be a meaningful change for you, even though both numbers fall squarely within the reference range.