Free Testosterone: What This Test Actually Measures

What free testosterone physically represents

Total testosterone counts every molecule of the hormone in a blood sample, whether it is bound to a carrier protein or floating free. That single number can obscure a clinically relevant detail: testosterone bound tightly to SHBG cannot cross cell membranes and is not available to activate androgen receptors while attached. A separate, loosely bound fraction on albumin can dissociate during capillary transit and become available; some clinicians combine that fraction with free testosterone into a "bioavailable testosterone" estimate. Free testosterone itself is the smallest slice: unbound, dissolved in serum, and immediately available to enter tissue and bind intracellular androgen receptors.

This is why two men can have very different clinical pictures despite similar total testosterone. A man with high SHBG can have a total testosterone that looks adequate while his free testosterone sits near the bottom of the range. A man with low SHBG (common with obesity and insulin resistance) can show a normal or even modest total testosterone while his free fraction runs high. Endocrine specialty guidance recommends measuring free testosterone specifically when total testosterone is borderline or when a condition known to distort SHBG is suspected, rather than as a routine first-line test in every patient.

Free testosterone measures a proportion, not an independent hormone. Anything that shifts total testosterone or SHBG shifts the free fraction along with it. Reading a free testosterone value in isolation, without knowing SHBG and the clinical context, is a common source of misinterpretation.

How the number is actually generated

The method used to produce a "free testosterone" result changes how much confidence you should place in it.

Equilibrium dialysis separates serum across a semipermeable membrane and measures the unbound testosterone that crosses it, typically by mass spectrometry. This is the reference method against which other approaches are validated, but it is slower and available mainly through reference laboratories.

Calculated free testosterone uses a formula (the Vermeulen equation is the most commonly cited) that estimates the free fraction from total testosterone, SHBG, and albumin. This approach is widely accepted as a practical substitute for equilibrium dialysis, though its accuracy is understood to degrade at the extremes of SHBG (very high or very low), where the assumptions built into the formula are less reliable.

Direct analog immunoassays use a labeled testosterone analog that competes for binding. These assays are fast and inexpensive but are considered the least reliable of the three methods; endocrine specialty guidance has cautioned against relying on direct analog free testosterone results for clinical decisions. If a lab report lists "free testosterone" without stating the method, it is reasonable to ask the lab or ordering clinician which assay produced it before treating the number as decision-grade.

A framework for interpreting a free testosterone result

Free testosterone cannot be read correctly in isolation. The table below is a starting framework, not a diagnostic tool, and it does not replace an in-person clinical evaluation.

The recurring exception across every row: a free testosterone number only means what it appears to mean once SHBG, the assay method, the time of draw, and the clinical symptoms are all accounted for together.

What the reference ranges are, and their real limits

Commonly cited reference ranges for free testosterone by equilibrium dialysis are approximately 35 to 155 pg/mL for adult men in their 20s through 40s, and roughly 0.1 to 6.4 pg/mL for premenopausal women. These figures are widely used by clinical laboratories, but they are not universal: every lab sets its own reference interval based on its assay and reference population, and a result should be interpreted against the range printed on that specific lab report rather than a number from an unrelated source.

Free testosterone is understood to decline with age in men, and more steeply than total testosterone, because SHBG tends to rise with age. In women, free testosterone is often the first androgen marker to move outside its normal range in polycystic ovary syndrome (PCOS), sometimes before total testosterone does. International PCOS guidance has favored calculated free testosterone (using formulas such as the Vermeulen equation) or mass-spectrometry-based methods over direct immunoassays for this reason.

A single low or high result is not, by itself, a diagnosis. Endocrine specialty guidance for suspected low testosterone in men calls for confirmation on at least two separate morning samples together with symptoms of androgen deficiency before a diagnosis of hypogonadism is made.

The SHBG connection: what moves the free fraction

Because free testosterone is a proportion, understanding what moves SHBG explains most of what moves free testosterone.

Conditions generally associated with higher SHBG (which lowers the free fraction relative to total testosterone) include aging, hyperthyroidism, liver disease such as cirrhosis, HIV infection, and use of estrogen-containing medications or certain anticonvulsants such as phenytoin or carbamazepine. A person on one of these drugs or with one of these conditions can have a total testosterone that looks adequate while their free testosterone is meaningfully lower.

Conditions generally associated with lower SHBG (which raises the free fraction relative to total testosterone) include obesity, insulin resistance and type 2 diabetes, hypothyroidism, nephrotic syndrome, and use of exogenous androgens or glucocorticoids. Obesity and insulin resistance are described in the endocrinology literature as the most common drivers of low SHBG encountered in practice.

None of these relationships are precise enough to predict an individual's free testosterone from their SHBG-affecting condition alone; they explain the direction of the shift, not the magnitude for any one person.

How free testosterone is used in testosterone therapy monitoring

For patients already on testosterone replacement therapy (TRT), free testosterone is commonly used alongside total testosterone to judge whether a dose is producing an adequate, but not excessive, level of bioavailable hormone. The goal in most protocols is not to maximize the number on the report but to place testosterone exposure within a physiologic range while symptoms improve and safety markers (such as hematocrit) stay in an acceptable range.

Two monitoring pitfalls are worth naming specifically because they are easy to miss when only total testosterone is checked:

• Apparent adequacy that is not real: a patient on TRT with high SHBG can show a total testosterone that looks generous while free testosterone sits at the low end, meaning the tissue-available dose may still be insufficient.
• Apparent moderation that is not real: a patient with obesity or low SHBG can show a total testosterone that looks unremarkable while free testosterone runs high, which is a setting where checking hematocrit and other safety markers becomes more important, not less.

Labs for TRT monitoring are generally drawn at trough (immediately before the next injection for injectable formulations) to capture the low point of the dosing cycle, since injectable TRT suppresses the body's own circadian testosterone rhythm. Morning timing matters less on TRT than it does for a diagnostic workup in someone not yet on therapy.

Free testosterone in women

Free testosterone is not only a PCOS marker. In premenopausal women, an elevated free testosterone alongside clinical signs of hyperandrogenism (acne, hirsutism, androgenic alopecia, menstrual irregularity) supports a PCOS evaluation, and international guidance has specified that calculated free testosterone or mass-spectrometry methods, not direct immunoassays, should be used for that purpose.

In postmenopausal women, free testosterone has been studied as one marker among several relevant to sexual desire and, separately, to bone health, though the evidence connecting testosterone levels to these outcomes in women is observational and less settled than the male hypogonadism literature. A consensus statement on testosterone therapy for women has noted that a single testosterone level does not reliably predict who will respond to treatment for low sexual desire, which argues against using free testosterone alone to decide who should or should not be offered therapy.

Common pitfalls that distort results

Wrong assay, high confidence. The most frequent error is treating a direct analog immunoassay result with the same confidence as a dialysis or calculated result. These assays can diverge substantially from equilibrium dialysis, and a clinically inconsistent result is a reason to ask about methodology before acting on it.

Wrong time of day. Testosterone secretion follows a circadian rhythm in men with an intact hypothalamic-pituitary-gonadal axis, generally peaking in the early morning and falling through the afternoon. A single afternoon draw in someone not on testosterone therapy can understate their true morning level and should not be used alone to diagnose deficiency; endocrine specialty guidance calls for fasting, morning (before roughly 10 a.m.) samples on two separate occasions for a diagnostic workup.

Biotin interference. The FDA issued a safety communication warning that high-dose biotin supplements (commonly found in hair, skin, and nail products, often at doses of several milligrams per day) can cause falsely high or falsely low results on immunoassay-based tests, including hormone testing, depending on the assay platform (a regulatory safety communication has warned of this interference). Discontinuing biotin supplementation for a period before testing, and telling the ordering clinician about biotin use, is a reasonable precaution.

What is established, what is plausible, and what is not settled

Established: Free testosterone is the unbound fraction of circulating testosterone. SHBG binds the large majority of circulating testosterone and can shift substantially with age, thyroid status, liver disease, obesity, insulin resistance, and certain medications, which changes the free fraction independent of total testosterone. Direct analog immunoassays are less reliable than equilibrium dialysis or calculated methods. Biotin can distort immunoassay results.

Plausible but not settled at the level of precise, generalizable numbers: How much any specific intervention (weight loss, resistance training, sleep, medications such as clomiphene) will change an individual's free testosterone, and by how much, varies across studies and populations; specific percentage changes cited in smaller trials should not be read as guarantees for an individual patient. The role of free testosterone specifically, versus total testosterone or clinical symptoms, in predicting who benefits from testosterone therapy in women is still an area of active discussion.

Not established from the material available here: Precise numeric relationships (for example, an exact percentage change in SHBG per unit of insulin resistance, or an exact percentage increase in free testosterone from a specific training protocol) require verification against the primary study before being repeated as a fixed figure. Where this draft references a general pattern from the literature without a verified citation, that pattern should be checked against the primary paper before being used in patient-facing material.

If a lab result is significantly out of range, unexpected given symptoms, or inconsistent with a prior result, that is a reason for a follow-up conversation with the ordering clinician rather than self-adjustment of any supplement, medication, or therapy dose. Symptoms such as new chest pain, difficulty breathing, or signs of a blood clot in someone on testosterone therapy warrant urgent evaluation rather than waiting for a routine follow-up lab.