This article explains the science of free testosterone calculation. It is not medical advice. Lab interpretation and TRT dosing decisions should be made with a licensed prescriber.
If you have ever stared at a TRT lab report and wondered why your "free testosterone" number is suspiciously low while your total testosterone looks normal, the answer is usually not your hormones. It is the assay your lab used. The direct-measurement free testosterone reported by Quest, LabCorp, and most other US commercial labs is known to be inaccurate -- consistently underestimating the actual unbound fraction by 30-50% relative to gold-standard equilibrium dialysis.
The fix is the calculator. The Vermeulen equation, published in 1999 and still endorsed by the Endocrine Society's testosterone therapy guideline, takes three numbers you can get reliably -- total testosterone, SHBG, and albumin -- and computes free testosterone from the equilibrium binding chemistry. It correlates with equilibrium dialysis at r = 0.918. The direct assay correlates at less than 0.5.
This guide covers how the calculation works, what numbers actually matter, why different calculators give different answers (it is almost always units), how to interpret your result for TRT decisions, and which clinics monitor free testosterone correctly.
Key Takeaways
- Free testosterone (~2% of total) is the biologically active fraction; calculated free T is more reliable than direct-assay free T
- The Vermeulen equation (1999) is the standard; correlates with equilibrium dialysis at r = 0.918
- You need three inputs: total testosterone, SHBG, and albumin (albumin defaults to 4.3 g/dL if missing)
- SHBG dominates the result; high SHBG = low free T even with normal total T
- Direct-assay free testosterone (most US labs) underestimates by 30-50% and should not drive clinical decisions
- Optimal TRT range: roughly 200-350 pg/mL calculated free T at trough
- Most online calculator differences are unit-conversion bugs (ng/dL vs nmol/L), not equation errors
What Free Testosterone Actually Is
Testosterone in your bloodstream exists in three pools:
- Bound to SHBG (sex hormone-binding globulin): Roughly 44-65% of total. This pool is biologically inactive -- SHBG holds testosterone tightly enough that it cannot dissociate fast enough to enter cells.
- Bound to albumin: Roughly 33-54% of total. This pool is loosely bound. Most physiologists and the Endocrine Society treat albumin-bound testosterone as bioavailable, because the binding is weak enough to release testosterone at the capillary level.
- Free (unbound): Roughly 1-3% of total. This is the fraction that diffuses freely into tissue and binds the androgen receptor.
"Bioavailable testosterone" = free + albumin-bound. "Free testosterone" = free only. Both are clinically useful; free testosterone is the more conservative measure and the one referenced in most TRT protocols.
The reason these pools matter is that total testosterone can be deceiving. A man with very high SHBG (say 80 nmol/L) and total T of 600 ng/dL will have almost no biologically active hormone available to his tissues despite a "normal" total. A man with very low SHBG (say 15 nmol/L) and total T of 350 ng/dL may feel and function like someone in the high-normal range. The total number alone does not tell you who is symptomatic.
Why You Cannot Just Measure Free Testosterone Directly
You can, technically. The reference method is equilibrium dialysis -- you put serum on one side of a membrane permeable only to free testosterone, let it equilibrate against a buffer, then measure how much testosterone crosses. It is accurate, slow, and expensive (typically $80-150, sent to specialty labs like Mayo or Quest's reference lab).
What most US commercial labs report when you order "free testosterone" is not equilibrium dialysis. It is the analog displacement direct assay -- a quick immunoassay that uses a labeled testosterone analog. The method has been criticized for over twenty years. The seminal critique by Rosner (2001) called it "an extraordinarily inaccurate assay." It systematically underestimates true free testosterone, particularly at the low and high ends of the range, and the bias varies by patient.
The practical consequence: if your TRT clinic ordered "free testosterone" without specifying equilibrium dialysis, the number on your report is the direct-assay number, and it should not drive a dose decision. Recalculate from total T and SHBG.
The Vermeulen Equation
Alex Vermeulen and colleagues at the University of Ghent published the calculation in 1999 in JCEM. The equation models the equilibrium between testosterone and its two binding proteins (SHBG and albumin) using known association constants. Inputs:
- Total testosterone (in nmol/L, or convert from ng/dL by dividing by 28.84)
- SHBG (in nmol/L)
- Albumin (in g/L, or convert from g/dL by multiplying by 10; defaults to 43 g/L if not measured)
Output: free testosterone in nmol/L (multiply by 28.84 to get ng/dL, or by 288.4 to get pg/mL).
The math involves solving a quadratic equation derived from the mass-action law -- not something most clinicians do by hand. Every reputable online calculator implements it. The original ISSAM web calculator (issam.ch/freetesto.htm) is the reference implementation; mdapp, Optimal DX, and Omnicalculator versions all produce the same answer when fed the same inputs in the correct units.
Worked Example
A 38-year-old man on TRT has a quarterly draw with:
- Total testosterone: 850 ng/dL (= 29.5 nmol/L)
- SHBG: 28 nmol/L
- Albumin: 4.4 g/dL (= 44 g/L)
Plug into the Vermeulen calculator:
- Free testosterone: roughly 0.62 nmol/L = 180 pg/mL
- Bioavailable testosterone: roughly 14.4 nmol/L = 415 ng/dL
His direct-assay free testosterone reported by Quest came back at 95 pg/mL -- 47% lower than the calculated value. The calculated number is the one to trust.
How SHBG Changes the Result
SHBG is the dominant variable. Hold total testosterone constant and watch what happens as SHBG moves:
| Total T |
SHBG |
Calculated Free T |
Interpretation |
| 600 ng/dL |
15 nmol/L |
~190 pg/mL |
Low SHBG -- robust free T despite mid-range total |
| 600 ng/dL |
30 nmol/L |
~125 pg/mL |
Average SHBG -- reference-range free T |
| 600 ng/dL |
60 nmol/L |
~75 pg/mL |
High SHBG -- functionally low free T |
| 600 ng/dL |
90 nmol/L |
~55 pg/mL |
Very high SHBG -- frank hypogonadal free T despite "normal" total |
This is why two men with identical total testosterone can have completely different symptom profiles. The man with high SHBG presents like classical hypogonadism -- low libido, fatigue, mood disturbance -- despite a total number that looks fine on paper.
The clinical takeaway is that SHBG should be drawn at every TRT visit, not just at baseline. SHBG can shift meaningfully with body composition changes, thyroid function, insulin sensitivity, and oral medications. If your last SHBG was 25 and your current is 45, your free T has fallen by roughly a third even if total T is unchanged. See the low SHBG TRT protocol article for the inverse pattern -- low SHBG patients who run high free T at modest total T.
How Albumin Changes the Result
Albumin matters less than SHBG but is not zero. The Vermeulen equation defaults to 43 g/L (4.3 g/dL) when albumin is not entered. Across a normal range of 3.5-5.0 g/dL, calculated free T varies by perhaps 10-15% in either direction. For most healthy adults the default is fine.
You should measure albumin if:
- You have liver disease (cirrhosis lowers albumin, raising free T artifactually if not measured)
- You have chronic kidney disease with proteinuria
- You are severely malnourished or hospitalized
- You take medications that bind to albumin in significant amounts
For routine TRT monitoring in a healthy outpatient, the default is acceptable.
Free vs Bioavailable -- Which Should You Track?
Both are valid. The difference is what you count as "active" testosterone:
- Free testosterone counts only the truly unbound fraction (~2%). More conservative. Endocrine Society uses this.
- Bioavailable testosterone counts free + albumin-bound (~35-55%). More inclusive. Some clinicians prefer this because albumin-bound testosterone does dissociate at the capillary level.
In practice the two track each other tightly because albumin varies less than SHBG. If you pick one and stick with it, the trend you see over multiple labs is what matters, not which definition you used. Most TRT prescribers use free testosterone.
What Numbers Are Optimal on TRT?
The reference ranges below assume Vermeulen-calculated free T from a sensitive (LC-MS) total testosterone assay. They are typical clinical targets, not regulatory cutoffs.
| Calculated Free T (pg/mL) |
Status |
| <50 |
Frank hypogonadal |
| 50-90 |
Below reference -- symptom likely |
| 90-150 |
Reference-range low |
| 150-250 |
Reference-range mid (typical untreated 30-40 yo) |
| 250-350 |
TRT optimal trough range |
| 350-500 |
TRT high-normal -- monitor for E2, hematocrit |
| >500 |
Supraphysiologic -- dose reduction warranted |
Trough values matter more than peak. A man on weekly cypionate may peak at 600 pg/mL day 2 and trough at 220 pg/mL day 7 -- the trough is what represents most of his week. See testosterone blood test timing for when to draw.
For untreated men investigating low T, the Endocrine Society guideline cites 64-70 pg/mL as the lower-bound diagnostic threshold for free testosterone deficiency in symptomatic men, in addition to a low total testosterone reading. Two morning draws below the threshold plus matching symptoms is the standard for hypogonadism diagnosis.
Why Online Calculators Sometimes Disagree
If two reputable calculators give different answers from the same inputs, the issue is almost always units, not formulas. Common conversions:
- Total testosterone: ng/dL × 0.0347 = nmol/L (or divide ng/dL by 28.84)
- SHBG: nmol/L is the standard everywhere -- usually no conversion needed
- Albumin: g/dL × 10 = g/L
- Free testosterone output: nmol/L × 28.84 = ng/dL, or nmol/L × 288.4 = pg/mL
US labs report total testosterone in ng/dL and free testosterone in pg/mL. European and most international labs use nmol/L throughout. If you see a free T of 0.6 in your result and another of 173 elsewhere, they may be the same number in different units (0.6 nmol/L ≈ 173 pg/mL).
A second smaller source of disagreement is the albumin association constant the calculator uses. Vermeulen used 3.6 × 10⁴ M⁻¹. Some calculators (e.g., the Mazer 2009 implementation) use slightly different constants based on more recent binding data, producing free T values 5-15% lower. The Fiers et al. 2018 reanalysis (PMID 29618085) addressed this. For clinical purposes the small offset does not change interpretation.
How to Use This in Your TRT Workup
A practical workflow:
- Order the right tests. Total testosterone (LC-MS preferred over immunoassay), SHBG, and albumin (or accept the default). Free testosterone direct assay can be added but treat it as a sanity-check, not the primary number.
- Calculate using Vermeulen. Use the ISSAM calculator or any reputable equivalent. Save the screenshot or write the number down.
- Compare to your prior labs. Free T trending matters more than the absolute value. A 30% drop quarter-over-quarter is signal even if the current number is "in range."
- Correlate with symptoms. Two morning draws under 70 pg/mL with libido, fatigue, mood, or strength symptoms supports hypogonadism. The same numbers without symptoms is a different conversation.
- Use it to titrate dose. Free T responds more linearly to dose changes than total T, especially in low-SHBG patients. A man whose total went from 700 to 950 with a dose increase but whose free T went from 220 to 330 is responding correctly. A man whose total moved similarly but free T barely budged probably has rising SHBG or a calculation issue worth investigating.
What Good TRT Clinics Do With Free Testosterone
A clinic that takes free T seriously will:
- Order SHBG and albumin at baseline and every quarterly draw, not just at intake
- Calculate free testosterone from Vermeulen rather than relying on direct-assay free T
- Trend free T over time alongside total T, hematocrit, and estradiol
- Adjust dose with reference to free T, especially in low- and high-SHBG patients
- Explain the result to the patient rather than reporting only the total number
A clinic that dismisses SHBG as "not necessary" or insists the direct-assay free T is the right number is not running the protocol with current evidence in mind. See how to choose a TRT clinic and questions to ask a TRT clinic for the broader monitoring criteria. The main clinic comparison page shows which providers include SHBG and free T in their standard panel.
Edge Cases and Common Pitfalls
Obesity and Insulin Resistance
Obesity lowers SHBG, often substantially. A man with BMI 35 and insulin resistance may run SHBG of 12-18 nmol/L. His total T may look low (say 380 ng/dL) while his calculated free T is unremarkable (~110 pg/mL). The "low T" diagnosis based on total alone may be incorrect. Weight loss frequently raises SHBG and total T together without changing free T much. See GLP-1 + testosterone in obese men for the metabolic interaction.
Aging
SHBG rises with age in most men. A 65-year-old with total T of 500 ng/dL and SHBG of 55 nmol/L has free T near 70 pg/mL -- borderline hypogonadal -- despite a total that looks adequate. The Endocrine Society guideline explicitly notes that older men with normal total T but low free T are candidates for therapy if symptomatic. See testosterone decline with age.
Anabolic Steroid Users
Supraphysiologic doses can crush SHBG to single digits. A bodybuilder on 600 mg/week of testosterone may have SHBG of 6 nmol/L and free T of 600+ pg/mL even though his total T is "only" 1500 ng/dL. The Vermeulen equation breaks down somewhat at SHBG extremes -- the result remains directionally correct but the absolute number is less precise.
Thyroid Disorders
Hyperthyroidism raises SHBG; hypothyroidism lowers it. If your SHBG is unexpectedly out of range, check TSH and free T4 before assuming the SHBG itself is the problem.
Medications That Bind Albumin
Phenytoin, salicylates at high dose, and warfarin compete for albumin binding sites. In men on these drugs the albumin-bound testosterone fraction may be displaced, raising apparent free T modestly. Rarely clinically relevant but worth noting.
Bottom Line
The free testosterone calculator is not optional for TRT. It is the standard. Total testosterone alone is misleading at SHBG extremes, and the direct-assay free T offered by US commercial labs is too inaccurate to drive dosing. The Vermeulen equation, fed by total T, SHBG, and albumin, gives a reproducible, validated number that tracks the gold-standard equilibrium dialysis at r = 0.918. Use it on every quarterly draw. Plot the trend. Use it to inform dose changes alongside symptoms.
The clinics that monitor free testosterone correctly order SHBG and albumin routinely, calculate Vermeulen, and explain the result. The clinics that do not are running an incomplete protocol -- and an incomplete protocol on TRT is the difference between symptom resolution and a year of guessing.
If you are not sure whether your current setup meets that bar, the TRT clinic comparison page walks through which providers include SHBG and calculated free T in their standard quarterly panel.
Related Reading
References
- Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab. 1999;84(10):3666-3672. PubMed
- Bhasin S, Brito JP, Cunningham GR, et al. Testosterone Therapy in Men With Hypogonadism: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2018;103(5):1715-1744. PubMed
- Fiers T, Wu F, Moghetti P, Vanderschueren D, Lapauw B, Kaufman JM. Reassessing Free-Testosterone Calculation by Liquid Chromatography-Tandem Mass Spectrometry Direct Equilibrium Dialysis. J Clin Endocrinol Metab. 2018;103(6):2167-2174. PubMed
- Rosner W. An extraordinarily inaccurate assay for free testosterone is still with us. J Clin Endocrinol Metab. 2001;86(6):2903. PubMed
- Ho CKM, Stoddart M, Walton M, Anderson RA, Beckett GJ. Calculated free testosterone in men: comparison of four equations and with free androgen index. Ann Clin Biochem. 2006;43(Pt 5):389-397. PubMed
- Morris PD, Malkin CJ, Channer KS, Jones TH. A mathematical comparison of techniques to predict biologically available testosterone in a cohort of 1072 men. Eur J Endocrinol. 2004;151(2):241-249. PubMed
- de Ronde W, van der Schouw YT, Pols HA, et al. Calculation of bioavailable and free testosterone in men: a comparison of 5 published algorithms. Clin Chem. 2006;52(9):1777-1784. PubMed
- Mulhall JP, Trost LW, Brannigan RE, et al. Evaluation and Management of Testosterone Deficiency: AUA Guideline. J Urol. 2018;200(2):423-432. PubMed