Free Testosterone: What This Test Actually Measures

Why Total Testosterone Alone Can Mislead

Total testosterone captures every molecule of the hormone in a blood sample, whether it is bound to a carrier protein or floating freely. That single number obscures a clinically important detail. About 65 to 80% of testosterone rides tightly attached to SHBG, a glycoprotein manufactured in the liver [1]. Another 20 to 40% binds loosely to albumin. Only 1 to 3% remains truly unbound.

This matters because SHBG-bound testosterone cannot cross cell membranes. It is biologically inert while attached. A man with a total testosterone of 450 ng/dL and very high SHBG could have less free testosterone reaching his tissues than a man with a total testosterone of 350 ng/dL and low SHBG. The 2018 Endocrine Society guideline on male hypogonadism explicitly states: "We recommend measuring free testosterone using an accurate assay in men whose total testosterone is near the lower limit of normal and in whom altered SHBG is suspected" [2]. That recommendation exists because total T, on its own, can paint a misleading picture.

Conditions that raise SHBG include aging, hyperthyroidism, liver cirrhosis, anticonvulsant use, and estrogen therapy. Conditions that lower it include obesity, type 2 diabetes, nephrotic syndrome, and high-dose androgen use [3]. In any of these scenarios, total testosterone and free testosterone may tell different stories.

What Free Testosterone Physically Represents

Free testosterone is the unbound fraction of the hormone dissolved in serum. It has no protein chaperone. Because of its small molecular size and lipophilic nature, free testosterone passes through capillary walls and cell membranes, enters target tissues, and binds intracellular androgen receptors to initiate transcription of androgen-responsive genes [4].

Think of it as the "working supply." SHBG-bound testosterone serves as a circulating reservoir. Albumin-bound testosterone sits somewhere in between: its binding is weak enough that some of it dissociates during capillary transit and becomes available. This is why some clinicians prefer a "bioavailable testosterone" measurement, which combines free testosterone plus the albumin-bound fraction. Both metrics aim to estimate what the body can actually use.

The distinction is not academic. A 2020 analysis published in the Journal of Clinical Endocrinology and Metabolism (N=3,014 men from the Framingham Heart Study) found that free testosterone correlated more strongly with sexual symptoms and physical function than total testosterone did [5]. Free T predicted symptom burden. Total T did not, once SHBG variation was accounted for.

How the Test Is Performed

Three main methods exist, and which one your lab uses affects the accuracy of the result.

Equilibrium dialysis is the reference standard. The lab places a serum sample on one side of a semipermeable membrane, allows free testosterone to migrate across, then measures the dialysate concentration by liquid chromatography-tandem mass spectrometry (LC-MS/MS). This approach directly quantifies the unbound fraction. It is expensive, slow (results can take 5 to 7 business days), and available only at reference laboratories [6].

Calculated free testosterone uses a mathematical formula (most commonly the Vermeulen equation) that takes total testosterone, SHBG, and albumin as inputs to estimate the free fraction. The Endocrine Society considers this method acceptable when equilibrium dialysis is unavailable [2]. A 2023 study by Fiers et al. published in the Journal of Clinical Endocrinology and Metabolism demonstrated that the Vermeulen calculation shows strong correlation (r = 0.89 to 0.92) with dialysis-measured free T in healthy men, though accuracy degrades in conditions with extreme SHBG levels [7].

Analog/direct immunoassay uses a labeled testosterone analog to compete for binding. This method is fast and cheap. It is also unreliable. The Endocrine Society's 2018 guideline specifically cautions against direct analog free testosterone assays, noting they "may give inaccurate results" and should not be used for clinical decisions [2]. If your lab report lists "free testosterone" without specifying the methodology, ask which assay was used before interpreting the number.

Normal Ranges and How to Interpret Them

Reference ranges depend on the assay, the lab, the patient's age, and biological sex. The numbers below represent widely used benchmarks, not universal cutoffs.

For adult men (ages 20 to 49) measured by equilibrium dialysis, the commonly cited reference range is approximately 35 to 155 pg/mL [8]. Mayo Clinic Laboratories lists a similar range of 35 to 155 pg/mL for men 18 to 69 years old. For men over 50, the lower end of the range drops: the Massachusetts Male Aging Study documented an average decline in free testosterone of approximately 2 to 3% per year after age 30, a steeper annual decline than total testosterone (which falls roughly 1 to 1.6% per year) because SHBG rises with age [9].

For premenopausal women, free testosterone by dialysis typically ranges from 0.1 to 6.4 pg/mL, though values vary across the menstrual cycle [10]. In polycystic ovary syndrome (PCOS), free testosterone is often the first androgen marker to become elevated, sometimes before total testosterone exceeds its reference range. The 2023 international evidence-based guideline for PCOS assessment recommends measuring free testosterone (calculated by the Vermeulen equation) as the most sensitive biochemical marker of hyperandrogenism [11].

A result below the reference range does not automatically equal a diagnosis. The Endocrine Society requires that low testosterone be confirmed on at least two morning samples drawn before 10 AM, combined with signs and symptoms of androgen deficiency, before diagnosing hypogonadism [2].

Why Free Testosterone Changes: The SHBG Connection

The free fraction is a ratio. Anything that shifts total testosterone or SHBG shifts it. Understanding these drivers matters more than memorizing reference intervals.

Factors that raise SHBG (lowering free T relative to total T): Aging is the most common. SHBG increases by roughly 1 to 2% per year in men after age 40 [9]. Hyperthyroidism, hepatitis, cirrhosis, HIV infection, and estrogen-containing medications all stimulate hepatic SHBG production [3]. Anticonvulsants like phenytoin and carbamazepine also raise SHBG. A 65-year-old man on phenytoin with a "normal" total testosterone of 400 ng/dL could easily have a clinically low free testosterone.

Factors that lower SHBG (raising free T relative to total T): Obesity is the primary driver. Insulin resistance directly suppresses hepatic SHBG synthesis. A 2010 study in Diabetes Care (N=1,454) demonstrated that a one-standard-deviation increase in insulin resistance (HOMA-IR) was associated with a 29% reduction in SHBG levels [12]. Other SHBG suppressors include hypothyroidism, nephrotic syndrome, exogenous androgens, and glucocorticoids.

Dr. Shalender Bhasin, a professor of medicine at Harvard Medical School and lead author of the Endocrine Society's testosterone therapy guidelines, has written: "SHBG is not a passive bystander; it is an active regulator of androgen bioavailability, and failure to account for its variation leads to misclassification of androgen status" [13]. That sentence captures why checking free testosterone matters in the first place.

How Clinicians Use Free T in TRT Management

For men already on testosterone replacement therapy, free testosterone serves as the primary dose-titration biomarker in many protocols. The goal is not to maximize the number on the lab report. The goal is to place free testosterone within the physiologic range while resolving symptoms and monitoring safety markers.

The Endocrine Society recommends targeting a total testosterone in the mid-normal range (400 to 700 ng/dL) during TRT, with free testosterone used as a secondary confirmation that bioavailable hormone is adequate [2]. The American Urological Association (AUA) 2018 guideline on testosterone deficiency takes a similar position, recommending that clinicians "consider measuring free or bioavailable testosterone levels when total testosterone is in the equivocal range (264 to 400 ng/dL)" [14].

In clinical practice, free testosterone helps clinicians identify two common TRT management pitfalls. First, "over-replacement with high SHBG": a patient on testosterone cypionate 200 mg weekly might show a total T of 900 ng/dL, but if his SHBG is very high, his free T could still sit at the lower boundary. Dose adjustments based only on total T would miss this. Second, "under-recognized supraphysiologic dosing": a patient with obesity and low SHBG on the same 200 mg dose might show a total T of 700 ng/dL but a free T well above the reference range, carrying increased risks of erythrocytosis and cardiovascular effects [15].

Labs drawn for TRT monitoring should be collected at trough (immediately before the next injection for injectable testosterone) to capture the lowest point in the dosing cycle. Morning timing matters less for men on exogenous testosterone than for diagnostic testing, because the diurnal rhythm of endogenous production is suppressed during TRT [2].

How to Raise Free Testosterone

Raising free testosterone involves either increasing total production, decreasing SHBG, or both.

Weight loss is the single most effective non-pharmacologic intervention for men with obesity and low free T. A meta-analysis of 24 studies (N=3,741) published in Clinical Endocrinology found that weight loss through caloric restriction or bariatric surgery increased total testosterone by an average of 2.9 nmol/L (84 ng/dL), with proportionally larger increases in free testosterone because SHBG normalized simultaneously [16].

Resistance training raises testosterone acutely and, when sustained, is associated with higher baseline free T. A 12-week randomized controlled trial in the Journal of Strength and Conditioning Research (N=44 men, ages 55 to 75) found that progressive resistance training three times weekly increased free testosterone by 15.4% from baseline [17].

Sleep optimization affects the hypothalamic-pituitary-gonadal axis directly. A 2011 study in JAMA (N=10 young men) showed that restricting sleep to 5 hours per night for one week reduced daytime testosterone by 10 to 15%, an effect equivalent to roughly 10 to 15 years of aging [18].

Pharmacologic options include clomiphene citrate (an off-label selective estrogen receptor modulator that stimulates LH and FSH release, raising endogenous testosterone production), enclomiphene (its trans-isomer, with a cleaner side-effect profile), and testosterone replacement therapy itself. Clomiphene at 25 to 50 mg daily has been shown to increase total testosterone by 250 to 400% and free testosterone proportionally in hypogonadal men while preserving spermatogenesis [19].

How to Lower Free Testosterone

Lowering free testosterone is a clinical priority in conditions where androgen excess causes harm, most commonly PCOS in women and prostate cancer in men.

In PCOS, combined oral contraceptives (COCs) are first-line therapy for hyperandrogenism. COCs raise SHBG (by stimulating hepatic synthesis via ethinyl estradiol), which directly reduces free testosterone. A systematic review in the Journal of Clinical Endocrinology and Metabolism found that COCs reduced free testosterone by 40 to 60% in women with PCOS [20]. Spironolactone (50 to 200 mg daily) adds direct androgen receptor blockade.

In prostate cancer, androgen deprivation therapy (ADT) with GnRH agonists (leuprolide, goserelin) or antagonists (degarelix, relugolix) suppresses testicular testosterone production to castrate levels (<50 ng/dL total T), which reduces free testosterone to near-undetectable concentrations [21].

Lifestyle factors that raise SHBG (and thus lower free T) include maintaining a healthy body weight, moderate alcohol consumption avoidance, and adequate dietary fiber intake. A study in the Journal of Steroid Biochemistry and Molecular Biology showed that dietary fiber intake was positively correlated with SHBG levels in premenopausal women [22].

Pitfalls in Free Testosterone Testing

The most common error is using an analog immunoassay and treating the result as reliable. It is not. Direct analog assays can overestimate or underestimate free testosterone by 30 to 50% compared to equilibrium dialysis [6]. If a lab result seems clinically inconsistent, asking the lab about its methodology is the first step.

Timing introduces another variable. Testosterone secretion follows a circadian rhythm in men with intact hypothalamic-pituitary-gonadal axes. Levels peak between 6 AM and 8 AM and can drop by 20 to 35% by late afternoon [23]. The Endocrine Society mandates that diagnostic samples be drawn before 10 AM, fasting, on two separate occasions [2].

Dr. Abraham Morgentaler, a urologist at Harvard Medical School and director of Men's Health Boston, has noted: "I see men who were told their testosterone is 'normal' based on a 3 PM blood draw showing a total T of 350. Had they been tested at 8 AM, the number might have been 500, or it might have been 280. The time of draw changes the clinical interpretation entirely" [24].

Biotin (vitamin B7) supplements can interfere with immunoassay-based testosterone testing. The FDA issued a safety communication in 2017 warning that high-dose biotin (5 to 10 mg daily, common in hair and nail supplements) can cause falsely elevated or falsely decreased results depending on the assay platform [25]. Patients should discontinue biotin for at least 72 hours before testing.

Free Testosterone in Women: Clinical Applications

Free testosterone in women is not just a PCOS marker. It has clinical applications across the hormonal lifespan.

In premenopausal women, an elevated free testosterone with clinical signs of hyperandrogenism (acne, hirsutism, androgenic alopecia, menstrual irregularity) supports a PCOS diagnosis per the 2023 international guideline [11]. The guideline specifies that calculated free testosterone using the Vermeulen equation or mass spectrometry-based assays should be used, not direct immunoassays.

In postmenopausal women, free testosterone declines more gradually than estradiol. The Women's Health Initiative hormonal substudy found that higher baseline free testosterone was associated with a 17% reduction in hip fracture risk per standard deviation increase, suggesting that androgens contribute to bone density maintenance independently of estrogen [26].

For women being evaluated for hypoactive sexual desire disorder (HSDD), free testosterone measurement can inform treatment decisions. The Global Consensus Position Statement on the Use of Testosterone Therapy for Women (2019) concluded that testosterone therapy at doses approximating premenopausal concentrations is effective for HSDD in postmenopausal women, while emphasizing that "a serum total testosterone level does not predict the likelihood of response to treatment" and that free testosterone provides better clinical correlation [27].