Free Testosterone
Calculator
Calculate your free and bioavailable testosterone from total T, SHBG, and albumin using the validated Vermeulen equation — the same method endocrinologists use.
| Free Testosterone | Interpretation |
|---|---|
| Below 5 ng/dL | Possibly low — often warrants clinical evaluation |
| 5 – 7 ng/dL | Low-normal — borderline, context-dependent |
| 7 – 21 ng/dL | Within typical range for adult men |
No universally validated age-stratified reference ranges exist specifically for the calculated Vermeulen method — these are general clinical threshold guides, not a substitute for your lab's own reference range and your doctor's interpretation.
What Is Free Testosterone and Why It Matters More Than Total T
This matters clinically because a normal total testosterone result can be genuinely misleading. A man with high SHBG can have a "normal" total T reading while his actual free testosterone — the biologically active portion — is low, producing real symptoms of testosterone deficiency despite normal-looking lab results. This is exactly why calculated free testosterone, not total testosterone alone, is considered the more clinically meaningful measurement in many evaluations.
The Vermeulen Equation: A Validated Method
Since direct measurement of free testosterone (via equilibrium dialysis) is expensive, technically demanding, and available at very few reference laboratories, the Vermeulen equation — published in 1999 by Alex Vermeulen, Lieve Verdonck, and Jean-Marc Kaufman in the Journal of Clinical Endocrinology and Metabolism — provides a validated mathematical alternative. Their landmark paper tested multiple calculation methods against the equilibrium dialysis gold standard and found the Vermeulen approach produced the strongest correlation (r > 0.9), using known binding-affinity constants for testosterone's interaction with SHBG and albumin, based on the law of mass action.
The Endocrine Society specifically recommends calculated free testosterone using this type of validated equation when direct equilibrium dialysis isn't available, while explicitly cautioning against analog free testosterone immunoassays (a cheaper, faster lab test type), which multiple studies have found to be unreliable and are not recommended for clinical use.
Why SHBG Levels Vary So Much
⬆️ Conditions That Raise SHBG
Aging, hyperthyroidism, liver disease, and estrogen use are all commonly associated with elevated SHBG — which lowers free testosterone even if total T stays the same.
⬇️ Conditions That Lower SHBG
Obesity, type 2 diabetes, hypothyroidism, and metabolic syndrome are all commonly associated with reduced SHBG — which can mean adequate free testosterone even at a relatively low total T reading.
Because SHBG shifts so much based on these common conditions, checking it alongside total testosterone — rather than relying on total T in isolation — gives a genuinely more accurate picture of your actual androgen status. This calculated free testosterone approach is particularly useful in obese men, older men, and anyone with a suspected SHBG abnormality, where total T alone is most likely to mislead.
Understanding Your Result
- Free testosterone below 5 ng/dL generally warrants clinical evaluation for possible hypogonadism, particularly alongside relevant symptoms
- 5-7 ng/dL is a borderline, low-normal range where clinical context and symptoms matter considerably in interpretation
- 7-21 ng/dL is generally within the typical range for adult men, though exact cutoffs vary by lab and calculation method
- No universally validated age-stratified ranges exist specifically for the Vermeulen calculated method — many online calculators apply age-based ranges derived from different assay types, which technically aren't directly interchangeable with calculated free T values
Limitations Worth Understanding
The Vermeulen equation, while validated and widely recommended, does have documented accuracy limitations worth knowing. Research has found the equation performs most reliably when SHBG concentrations fall within a moderate range (roughly above 30 nmol/L), with accuracy decreasing somewhat outside this range — a genuine limitation shared by essentially all calculated free testosterone methods, not unique to Vermeulen specifically. Additionally, the equation assumes standard, unaltered binding characteristics for SHBG and albumin; certain conditions (including some genetic SHBG variants and severe illness affecting albumin production) can theoretically affect accuracy, though this is a relatively uncommon consideration for most routine evaluations. As with any calculated estimate, results should be interpreted as a well-validated approximation rather than a perfectly precise direct measurement.
When to Get Tested and What Affects Results
Testosterone follows a natural daily rhythm, peaking in the early morning and gradually declining through the day, which is why blood draws for testosterone testing are conventionally recommended in the morning (typically before 10 AM) for the most standardized, comparable result. Because testosterone can also fluctuate meaningfully day-to-day and in response to acute illness, poor sleep, or significant stress, a single low result — particularly one close to a borderline threshold — is often worth confirming with a repeat morning test before drawing firm conclusions, which is standard practice many endocrinology guidelines specifically recommend.
When This Test Is Most Useful
Calculated free testosterone is particularly valuable in specific clinical scenarios: evaluating possible hypogonadism when total testosterone falls in a borderline range (roughly 200-400 ng/dL), assessing androgen status in obese or older men where SHBG commonly shifts, monitoring testosterone replacement therapy, and evaluating androgen excess in women (including as part of PCOS assessment). If total testosterone is either clearly normal or clearly low on its own, calculated free T often adds less additional diagnostic value — it's most useful specifically in the ambiguous middle range.
Frequently Asked Questions
Related Health Calculators
