How to Read a Testosterone Panel

This page is about the sheet in your hand, not about the calculations. Converting units, running the Vermeulen equation, placing a result against the published centiles by age and looking up which analytes have a sourceable figure are all done elsewhere on this site and the links at the top go straight to them. What none of those pages does is tell you what to look at first, and that turns out to be where most misreadings happen — not in the arithmetic, but in reading line three before line one.

Two habits cause most of the trouble. The first is reading a result before reading the conditions it was collected under: a testosterone drawn at four in the afternoon, after lunch, in the week after a chest infection is a measurement of an afternoon rather than of a man. The second is reading a calculated free testosterone without looking at the SHBG it was calculated from, which is reading an output without its input. Both are fixed by order rather than by knowledge.

One thing this page is blunt about throughout. Most of the lines on a men’s testosterone panel have no reference range that can honestly be printed, and the reasons split into two kinds that must not be confused. For calculated free testosterone, bioavailable testosterone and estradiol in men, this site went looking and found that none has been published — so no row on our calculators is marked normal. For SHBG, LH, FSH, prolactin and haematocrit, this site simply has not done that work, which is a statement about us and not about the literature. The first table keeps the two apart, and it is generated from the same module the printable reference sheet is built from so the two cannot drift.

What this page does not cover

These live elsewhere on the site and are not restated here.

• Lab reference range chart

  Which analytes have a sourceable figure at all, what kind of number each one is, and the eight ranges that are named and refused. This page uses that module’s verdicts rather than restating them.

• Free testosterone calculator

  Runs the Vermeulen equation on your own total testosterone, SHBG and albumin and returns free, bioavailable, albumin-bound and SHBG-bound fractions. This page computes nothing from your numbers.

• Bioavailable testosterone calculator

  The same equation reported as bioavailable testosterone, with the free, albumin-bound and SHBG-bound split, and the two albumin molar masses printed side by side.

• Testosterone unit converter

  ng/dL to nmol/L and back for total, and pg/mL to ng/dL to pmol/L for free, including the spread produced by the printed 0.0347 factor against the exact one.

• Testosterone reference range by age

  Places a total testosterone against the published age-stratified centiles for men 19 to 99, and shows what age-adjusting is and is not worth.

• Testosterone levels by age chart

  The printable version of the same figures, with the assay standardisation table and the decade rows this site declined to print because their provenance could not be traced.

• Male hypogonadism

  The diagnostic criteria themselves, quoted from both anchor guidelines, including the two places they contradict each other. This page reads a panel; it does not diagnose from one.

• What is in a men’s lab panel

  What a panel measures analyte by analyte across the whole sheet, and the contents comparison this site declined to draw because no provider discloses one.

The procedure

1. 1

   Read the collection details before you read a single result

   Find the draw time and whether you were fasting, and put them next to the result. The guideline does not ask for a testosterone; it asks for a fasting morning testosterone, and a sample that was neither is not the measurement the criteria are written against. Add two more things from your own memory: whether you were unwell or recovering that week, and whether you were on anything that suppresses testosterone. If any of those is wrong, the honest reading of the whole sheet is that it needs repeating rather than interpreting.

   Published trial

   “We recommend measuring fasting morning total T concentrations using an accurate and reliable assay as the initial diagnostic test.”

   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. PMID 29562364 — abstract. Only the abstract is open; the publisher full text is behind a wall, which is also why the third column of the first table says "not named in the abstract" rather than "not recommended".
2. 2

   Count how many times this has actually been measured

   One result is a measurement and two are a finding. Both major guidelines require the confirmation and the American one makes it a Strong recommendation at Grade A, specifying that both draws are early morning and on separate occasions. This is the single most useful question to ask of a sheet that has been used to justify a prescription, because a diagnosis made on one draw has not met either guideline’s minimum — and roughly a third of first low results do not reproduce.

   Public-health guidance

   “The diagnosis of low testosterone should be made only after two total testosterone measurements are taken on separate occasions with both conducted in an early morning fashion.”

   Mulhall JP, Trost LW, Brannigan RE, et al. Evaluation and Management of Testosterone Deficiency: AUA Guideline. J Urol. 2018;200(2):423–432. Statements 2, 3, 6, 7, 9, 11, 12, 22 and 30 are quoted on this page with their strength and evidence grades.
3. 3

   Read total testosterone against the published distribution, not only the printed range

   The interval beside your result belongs to that laboratory’s assay and sampled population and is not the diagnostic threshold. The one figure on the whole panel that is a reference interval in the proper sense comes from Travison 2017, which pooled four cohorts and re-measured a subset of each at the CDC to put them on one scale. Read your number against that distribution as well as the printed one, and note where the two disagree — a result can sit inside a laboratory range and below a guideline threshold at the same time.

   Published trial

   “In healthy nonobese men, 19 to 39 years, harmonized 2.5th, 5th, 50th, 95th, and 97.5th percentile values were 264, 303, 531, 852, and 916 ng/dL, respectively.”

   Travison TG, Vesper HW, Orwoll E, et al. Harmonized Reference Ranges for Circulating Testosterone Levels in Men of Four Cohort Studies in the United States and Europe. J Clin Endocrinol Metab. 2017;102(4):1161–1173. PMID 28324103 — the one figure on this panel that is a reference interval in the proper sense.
4. 4

   Read SHBG before you read anything called free

   SHBG is the line that tells you whether the total can be taken at face value. The guideline asks for a free testosterone specifically when the total sits near the lower limit or when something has altered SHBG — ageing, obesity, diabetes, thyroid disease, liver disease and several drugs all do. If your sheet has a free testosterone and no SHBG, you are holding an output with its input missing. The third table on this page shows how far a single total can move once SHBG is allowed to vary.

   Published trial

   “In men whose total T is near the lower limit of normal or who have a condition that alters sex hormone-binding globulin, we recommend obtaining a free T concentration using either equilibrium dialysis or estimating it using an accurate formula.”

   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. PMID 29562364 — abstract. Only the abstract is open; the publisher full text is behind a wall, which is also why the third column of the first table says "not named in the abstract" rather than "not recommended".
5. 5

   Find out which method produced the free testosterone

   There are three and they are not interchangeable. Equilibrium dialysis is the reference method. A calculation from total testosterone, SHBG and albumin — most often the Vermeulen equation — is the accepted alternative and is what our calculator runs. A direct analog immunoassay is the cheap one many panels include, and the Vermeulen paper this site implements found it unreliable in the original comparison — its values were only a fraction of the dialysis result, and by a fraction that itself moved with SHBG. If the report does not say which was used, that is a question for the laboratory rather than an assumption to make. Whichever it was, there is no harmonized reference range for a calculated free testosterone, so read it against the interval your own laboratory prints for its own method.

   Published trial

   “Values for aFT, however, were only a fraction of either AFTC or FT, the fraction varying as a function of SHBG levels. Also, the FAI/AFTC ratio varied as a function of the SHBG levels, and hence, neither aFT nor FAI is a reliable index of bioavailable T.”

   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. PMID 10523012 — the equation implemented in src/lib/free-testosterone.ts and used to compute the third table on this page.
6. 6

   Read LH next, because it says where the problem is

   This is the line most often missing from a direct-to-consumer panel and the one that changes the investigation most. A high LH alongside a low testosterone points at the testes. A low or inappropriately normal LH points at the pituitary or hypothalamus, and that branch leads to prolactin, iron studies and sometimes imaging. The American guideline makes measuring it a Strong recommendation at Grade A, and adds prolactin as a Strong recommendation at Grade A specifically when LH is low or low-normal.

   Public-health guidance

   “In patients with low testosterone, clinicians should measure serum luteinizing hormone levels.”

   Mulhall JP, Trost LW, Brannigan RE, et al. Evaluation and Management of Testosterone Deficiency: AUA Guideline. J Urol. 2018;200(2):423–432. Statements 2, 3, 6, 7, 9, 11, 12, 22 and 30 are quoted on this page with their strength and evidence grades.
7. 7

   Look for estradiol only if there is a reason it belongs there

   Estradiol appears on a great many commercial men’s panels and appears in the guidelines in one narrow situation. The American guideline asks for it where a man with testosterone deficiency presents with breast symptoms or gynecomastia, before starting therapy — that is the indication, and it is Expert Opinion rather than a graded recommendation. If it is on your sheet anyway, note that no reference range for estradiol in men has been established that this site could source, so a value flagged high or low by the laboratory is being flagged against that laboratory’s own interval.

   Public-health guidance

   “Serum estradiol should be measured in testosterone deficient patients who present with breast symptoms or gynecomastia prior to the commencement of testosterone therapy.”

   Mulhall JP, Trost LW, Brannigan RE, et al. Evaluation and Management of Testosterone Deficiency: AUA Guideline. J Urol. 2018;200(2):423–432. Statements 2, 3, 6, 7, 9, 11, 12, 22 and 30 are quoted on this page with their strength and evidence grades.
8. 8

   Treat haematocrit and PSA as pre-treatment lines rather than diagnostic ones

   Neither tells you anything about whether you have testosterone deficiency; both can stop you being treated for it. The American guideline requires haemoglobin and haematocrit before therapy is offered, with the polycythaemia risk explained, and makes a PSA in men over 40 a clinical principle before starting. Read them as gates rather than as findings, and read a PSA in particular as a number that will be harder to interpret later if nobody has a baseline.

   Public-health guidance

   “Prior to offering testosterone therapy, clinicians should measure hemoglobin and hematocrit and inform patients regarding the increased risk of polycythemia.”

   Mulhall JP, Trost LW, Brannigan RE, et al. Evaluation and Management of Testosterone Deficiency: AUA Guideline. J Urol. 2018;200(2):423–432. Statements 2, 3, 6, 7, 9, 11, 12, 22 and 30 are quoted on this page with their strength and evidence grades.
9. 9

   If you are already on treatment, read a different sheet entirely

   The reading order above is for a man being assessed. On treatment the question changes from what is wrong to whether the target has been reached and what has moved with it, and the interval changes too. The American guideline sets the monitoring cadence at every six to twelve months, asks for an initial follow-up level after an appropriate interval, and gives an explicit dosing target rather than leaving it to a range on a report. Doses and formulations are not on this page; the dosage chart linked below carries the labelled figures.

   Public-health guidance

   “Testosterone levels should be measured every 6-12 months while on testosterone therapy.”

   Mulhall JP, Trost LW, Brannigan RE, et al. Evaluation and Management of Testosterone Deficiency: AUA Guideline. J Urol. 2018;200(2):423–432. Statements 2, 3, 6, 7, 9, 11, 12, 22 and 30 are quoted on this page with their strength and evidence grades.

Every line on the panel, and what it can be read against

Generated from src/lib/lab-panels.ts, the same record /charts/lab-reference-range-chart is built from, so the two cannot disagree. Read the second column first: of the ten lines here, one has a reference interval in the proper sense and one has a decision threshold. The other eight have no number printed, for two different reasons.

Two different refusals are in that second column and collapsing them would be the commonest error on a page like this. "We looked; none is published" is a claim about the literature — it is what this site found for calculated free testosterone, bioavailable testosterone and estradiol in men, and it is why no row on our calculators is marked normal. "Not sourced on this site" is a claim about us — it applies to SHBG, LH, FSH, prolactin and haematocrit, and it must never be read as meaning no interval exists. Read those five against the range your own laboratory prints. The third column is limited in a way worth naming: only the abstract of the Endocrine Society guideline was opened for the source module, so "not named" means not named in the abstract.

The order to read it in, and what each line changes

A panel is not a list of independent numbers. Read it in this order and each line constrains how you read the next one; read it out of order and a normal total testosterone stops the process before SHBG has had a chance to explain the result.

Step 4 before step 5 is the one people invert. A calculated free testosterone is derived from the total and the SHBG, so reading it without looking at the SHBG that produced it is reading an output without its input. The third table shows how much difference that input makes.

The same total testosterone at five different SHBG values

Computed by calculateFreeTestosterone in src/lib/free-testosterone.ts — the same tested Vermeulen implementation /tools/free-testosterone-calculator uses, at the default albumin of 4.3 g/dL. This is arithmetic demonstrating a point, not a reference table: no cell here is marked normal or low, because no harmonized reference range for calculated free testosterone exists to mark it against.

Read down a single column. The total testosterone is identical in every cell of that column and the calculated free result changes by a factor of roughly three across the SHBG range shown. That is the argument for ordering SHBG whenever a total is being interpreted, and it is also the reason a calculated free value quoted without its SHBG tells you very little. The SHBG values here are a modelling grid chosen to span what laboratories commonly report, not a reference range; neither the grid nor any output cell is a normal value.

When it does not go to plan

What we could not source

Widely taught techniques we went looking for and could not anchor to a label, a public-health body or a trial. Each entry names what was searched and what this page does instead.

Frequently asked questions

What time of day should a testosterone test be done?

Morning, and both guidelines say so. The Endocrine Society recommends "measuring fasting morning total T concentrations using an accurate and reliable assay as the initial diagnostic test", and the American Urological Association requires both measurements to be taken "in an early morning fashion" as a Strong recommendation at Grade A. The size of the diurnal swing is what makes it matter: the AUA reports that in men aged 30 to 40 afternoon values run 20–25% below morning ones, with a much smaller difference in older men. An afternoon draw can move a result across a threshold without anything about the man changing.

Why does my report say normal when the guideline says I am low?

Because they are two different yardsticks. The interval printed beside your result is that laboratory’s own, derived from its assay and its sampled population — which is why the same blood can read normal at one laboratory and low at another. The guideline thresholds are fixed numbers: 264 ng/dL for the Endocrine Society, 300 ng/dL for the AUA, with the AUA noting that medical societies have used thresholds ranging from 230 to 350 ng/dL. The one true reference interval this site prints is Travison’s 264 to 916 ng/dL, from healthy nonobese men aged 19 to 39 on a CDC-calibrated assay. Read your number against both and treat a disagreement between them as information rather than as an error.

What is a normal free testosterone?

Nobody has published a figure this site is willing to quote. There is no harmonized reference range for calculated free testosterone, and no CDC standardisation programme for one, which is why no row on our free testosterone calculator, our unit converter or our printable reference sheet is marked normal. What can be said is how much it moves with SHBG: at a fixed total testosterone, the calculated free result changes by roughly a factor of three across the SHBG range laboratories commonly report — which is the third table on this page. Read your own value against the interval your laboratory prints for its own equation.

Do I need SHBG measured?

If a free testosterone is being interpreted, yes, because the free value is calculated from the total and the SHBG and reading it without the SHBG is reading an output with its input hidden. The Endocrine Society’s trigger for obtaining a free testosterone at all is a total "near the lower limit of normal" or a condition that alters SHBG — and the list of things that alter it is long: ageing, obesity, diabetes, thyroid disease, liver disease, HIV and several drug classes. Note the separate point in the first table: this site has not sourced a reference interval for SHBG itself, which is a statement about our work rather than about the literature.

How many testosterone tests do I need before anything is decided?

Two, and both in the early morning. The AUA makes it a Strong recommendation at Grade A: "The diagnosis of low testosterone should be made only after two total testosterone measurements are taken on separate occasions with both conducted in an early morning fashion." The Endocrine Society requires the same confirmation. Neither specifies how far apart the two draws should be, and the AUA says why — there is no definitive evidence on the optimal interval. The reason the repeat matters is that around 30% of men with an initial result in the hypogonadal range are normal on the second sample.

What is missing from most direct-to-consumer testosterone panels?

LH, most often, and it is the line that decides where the investigation goes next — high LH points at the testes, low or inappropriately normal LH at the pituitary or hypothalamus. The AUA makes measuring it a Strong recommendation at Grade A, and adds prolactin as a Strong recommendation at Grade A when LH is low or low-normal. SHBG is the second common omission, and without it a free testosterone cannot be interpreted. A panel that reports a single total testosterone has measured the thing that is easiest to sell and has not measured the two things that would tell you what it means.

Sources

1. [1]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. PMID 29562364 — abstract. Only the abstract is open; the publisher full text is behind a wall, which is also why the third column of the first table says "not named in the abstract" rather than "not recommended".
2. [2]Mulhall JP, Trost LW, Brannigan RE, et al. Evaluation and Management of Testosterone Deficiency: AUA Guideline. J Urol. 2018;200(2):423–432. Statements 2, 3, 6, 7, 9, 11, 12, 22 and 30 are quoted on this page with their strength and evidence grades.
3. [3]Travison TG, Vesper HW, Orwoll E, et al. Harmonized Reference Ranges for Circulating Testosterone Levels in Men of Four Cohort Studies in the United States and Europe. J Clin Endocrinol Metab. 2017;102(4):1161–1173. PMID 28324103 — the one figure on this panel that is a reference interval in the proper sense.
4. [4]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. PMID 10523012 — the equation implemented in src/lib/free-testosterone.ts and used to compute the third table on this page.
5. [5]Quest Diagnostics Clinical Education Center, FAQ 165 — Testosterone Testing. Quoted in the troubleshooting section on assay method: immunoassay "produces reliable results for normal or elevated levels of TT but lacks sensitivity, specificity, and precision at lower concentrations", and "Direct measurement of FT and bioavailable T is not available, so concentrations are estimated from calculations." A commercial laboratory’s own education material, named as such rather than treated as a guideline.