TL;DR: Low testosterone is diagnosed with a symptom pattern plus a low morning, fasting total testosterone draw, run on an accurate assay, backed by free testosterone calculated from SHBG, and paired with LH and FSH to find out why it’s low. That’s the standard the Endocrine Society and the American Urological Association actually publish. It’s almost never what happens. What happens is one draw at 4 PM, after lunch, total testosterone only, compared against a reference range built from a sick population. Every one of those shortcuts pushes the number in the same direction, and none of them tells you the cause. A single number is a screening. A diagnosis is a workup. If you were told you’re “normal” off one afternoon blood draw, you didn’t get a testosterone evaluation. You got a coin flip.

How Is Low Testosterone Actually Diagnosed?

Symptoms, plus low morning fasting draws, plus the free testosterone, plus the labs that explain the cause. Four parts. Not one number.

Here’s what both major guidelines say, in plain English. The Endocrine Society’s 2018 clinical practice guideline (Bhasin et al., JCEM) says you diagnose testosterone deficiency in men with both consistent symptoms and unequivocally low morning total testosterone, confirmed on a repeat measurement, using a reliable assay. The AUA’s 2018 guideline (Mulhall et al., J Urol) says the same thing and puts a number on it: two total testosterone measurements, both drawn early in the morning, both below roughly 300 ng/dL. We feel this is too low a barrier often times.

Read that again. Morning. Free Testosterone. Confirmed. Then, once you know it’s real, you find out why it’s happening, because “low” is a finding, not a diagnosis. Low because your testicles stopped producing is a different disease than low because your pituitary went quiet, and those get worked up differently.

Now compare that to what your physical actually did. One tube of blood, drawn whenever you happened to have the appointment, probably after you ate, run on a cheap assay, total testosterone only, no LH, no SHBG, no free T. Then the result gets held up against a range and you get a thumbs up.

That isn’t a lesser version of the workup. It’s a different thing wearing the workup’s clothes.

Why Does the Time of Day Matter So Much?

Because testosterone runs on a daily clock, and your afternoon number can be meaningfully higher or lower than the number that actually diagnoses you.

Testosterone follows a circadian rhythm. It peaks in the early morning and drifts down across the day. This isn’t a fringe finding, it’s the reason both guidelines specify a morning draw in the first place, generally before 10 AM. The entire diagnostic threshold you’re being measured against was built on morning values. Draw at a different time and you’re comparing your number to a standard it was never meant to be compared to.

That cuts both ways, and this is the part nobody explains.

If you draw in the afternoon, a man who is genuinely low can land close enough to the line to get waved through. And a man who is fine can occasionally look low and get worked up for a problem he doesn’t have. Bad timing doesn’t just hide low testosterone. It adds noise to a number that gets treated like scripture.

So when your result came back at 340 from a 3:45 PM draw, the honest read isn’t “you’re normal.” The honest read is “we don’t know yet.”

Why Do You Have to Be Fasting?

Because eating, and specifically a glucose load, drops your testosterone measurably within the hour.

This is one of the cleanest findings in the field and one of the most ignored. Caronia and colleagues (Clin Endocrinol, 2013) gave men a standard 75-gram oral glucose load and measured what happened. Total testosterone fell roughly 25 percent on average, and stayed down for the next two hours. A meaningful slice of the men who started with normal testosterone dropped into the hypogonadal range on the spot, purely from the sugar. The title of the paper says the quiet part out loud: implications for screening for hypogonadism.

Sit with that. A sandwich and a Coke at noon can move your 1 PM testosterone by a quarter of its value. That’s not a rounding error. That’s the difference between “you’re fine” and “let’s look closer,” delivered by your lunch.

So: morning, fasted. Not because we’re a prude about it. Because the number is fragile and everybody treats it like it’s carved in stone.

What Number Actually Counts as Low?

The AUA uses roughly 300 ng/dL of total testosterone as a reasonable diagnostic cut-off, and the widely used “264 to 916” range comes from Travison et al. (JCEM, 2017). Both are more fragile than they look.

The 264 to 916 range gets quoted like it descended from a mountain. It’s a harmonized reference range built from four population cohorts, and here’s the thing about population cohorts: they’re full of the population. Overweight men, poorly slept men, insulin-resistant men, men on medications that flatten testosterone. The range describes them. Then it gets used to dismiss you. That argument deserves its own space and it has one, because that range was built from a population that was never healthy to begin with.

The cut-off has a second problem, and it’s arguably worse: total testosterone is not the number that determines how you feel. Most of your testosterone is bound to SHBG and unavailable to your tissues. The fraction that’s actually free and doing work is what matters, and it can be deficient while your total sits comfortably “in range.” That’s free testosterone, calculated correctly, and our optimization target is 20 to 25 ng/dL.

So a real diagnosis doesn’t hang on one threshold. It reads the total, the SHBG, the calculated free T, and the man sitting in front of you, together. A number without a symptom pattern isn’t a diagnosis. A symptom pattern without the right numbers isn’t either.

Why Can’t the Lab Just Measure Free Testosterone Directly?

It can, and most of the time it does it badly. The Endocrine Society specifically advises against the direct analog free testosterone immunoassays that commercial labs love, because they’re unreliable.

There are two ways to get a trustworthy free testosterone. Equilibrium dialysis, which is the reference method and is neither fast nor cheap. Or calculation, using your total testosterone, SHBG, and albumin run through the Vermeulen equation, published by Belgian endocrinologist Alex Vermeulen in JCEM in 1999. The math is ugly and every modern calculator handles it instantly.

What most labs hand back instead is a direct analog assay, which is quick, cheap, and wrong often enough that the guideline tells you not to use it.

The same goes for total testosterone. Assays vary, which is why the guidelines specify an accurate, reliable, standardized assay (the CDC runs a hormone standardization program for exactly this reason). Mass spectrometry is the gold standard.

Translation: the number can be wrong before anyone even interprets it. Timing, fasting, one draw, a bad assay. Four independent chances to get a wrong answer, stacked on top of each other, presented as medical care.

This is what Apex’s panels are built to do properly: morning fasted draws, total and SHBG measured so free testosterone is calculated rather than guessed at, and the surrounding markers that explain the result. Order your panel or book a consult here.

Why Are LH and FSH the Two Most Skipped Labs?

Because they answer the only question that changes what happens next: why is it low. And they cost almost nothing.

Low testosterone splits into two categories, and telling them apart takes two labs:

• Primary (the testicles): testosterone is low, LH and FSH are high. The pituitary is shouting and the testicles aren’t answering.
• Secondary (the pituitary or hypothalamus): testosterone is low, LH and FSH are low or inappropriately normal. The signal upstream went quiet.

That distinction is not academic. Secondary hypogonadism has causes you genuinely need to find, and some of them are the actual health story of that man’s life:

• A prolactin-secreting pituitary tumor
• Hemochromatosis (iron overload)
• Obstructive sleep apnea
• Obesity and insulin resistance
• Opioids, glucocorticoids, and other medications
• Prior anabolic steroid use

The guidelines follow the thread: check prolactin, consider iron studies, and image the pituitary when secondary hypogonadism is severe (total T under roughly 150 ng/dL), when prolactin is persistently high, or when there are other pituitary hormone deficits or mass-effect symptoms.

A clinic that hands a man testosterone without ever measuring an LH has skipped the part where you find out what’s wrong with him. Sometimes what’s wrong is a tumor. Usually it’s his metabolic health. Either way, he deserved to know, and testosterone alone would have masked the symptom while the cause kept running.

What Else Belongs in a Real Workup?

Everything that either causes low testosterone, imitates it, or changes whether therapy is safe. The hormone is one line on the page.

This is where a screening and an evaluation stop resembling each other. The labs that belong:

• Total testosterone (morning, fasted, accurate assay)
• SHBG and albumin (so free T is calculated, not estimated)
• LH and FSH (primary vs. secondary, the cause)
• Prolactin (pituitary)
• Estradiol (sensitive assay, the other half of the hormonal picture)
• TSH and thyroid panel (the most common impostor)
• CBC with hematocrit (a baseline before anything, and hematocrit at or above 54% is where we recommend donating blood)
• Comprehensive metabolic panel (liver and kidney function)
• HbA1c and fasting insulin (insulin resistance drives low testosterone and gets missed for years)
• Ferritin and iron studies (causes fatigue in its own right)
• Vitamin D, 25-OH (target 60 to 80 ng/mL)
• ApoB (the cardiovascular marker that actually predicts risk, target under 80)
• PSA where age-appropriate

Notice how much of that list has nothing to do with testosterone. That’s the point. Half the men who show up convinced they have low T have something else, or have low T because of something else. The symptom pattern that gets written off as age overlaps almost perfectly with thyroid disease, sleep apnea, insulin resistance, iron deficiency, and depression. If nobody checked, nobody knows.

And the other half of the list exists because if therapy ever does become appropriate, you need to know where the man started. A baseline hematocrit taken after treatment begins is not a baseline. It’s a guess with a date on it.

What a Real Diagnosis Looks Like

A 46-year-old attorney. Trains four days a week, sleeps badly, up about 20 pounds over a decade. Tired by 2 PM, libido gone, morning erections rare enough that he noticed they stopped. He asks his doctor to check his testosterone.

Draw is at 3:50 PM, an hour after a turkey sandwich. Total testosterone comes back 335. In range. “Your testosterone is normal, you’re just busy. Try to get more sleep.”

He comes to us. Same man, different protocol. 7:45 AM, fasted: total testosterone 246. SHBG 58. Albumin 4.4. Free testosterone, calculated correctly: 4.6 ng/dL, against a target of 20 to 25. LH 2.1, so not his testicles, the signal upstream is quiet. Prolactin normal. Pituitary workup unremarkable. HbA1c 5.9. Fasting insulin 18.

There’s the whole story. He has confirmed testosterone deficiency, it’s secondary, and it’s being driven by the metabolic dysfunction nobody had measured. His afternoon number was not “normal.” It was 90 points of circadian rhythm and lunch, sitting on top of a real problem.

His doctor was technically correct and clinically wrong. And the fix for the man wasn’t a prescription pad. It started with the finding nobody looked for.

Screening vs. Evaluation: The Actual Difference

One tells you a number. The other tells you what’s happening to you.

• A screening: one draw, any time of day, non-fasted, total testosterone only, direct free T assay if any, compared to a population range, interpreted in 30 seconds.
• An evaluation: symptoms documented, morning fasted draws on an accurate assay, SHBG and albumin so free T is calculated, LH and FSH to establish the cause, the mimics ruled out, baselines captured, and a provider who tells you what it means.

The gap between those two is where a decade of a man’s life goes.

What to Do If You’ve Already Been Told You’re Fine

Ask one question: when was the draw, and what was on the panel?

If the answer is “afternoon,” or “after lunch,” or “just total testosterone,” you have not been evaluated. You’ve been screened, and the screening had at least one thermodynamic thumb on the scale. That doesn’t automatically mean you have low testosterone. It means the test that told you that you don’t wasn’t built to answer the question.

Go get the real version. Morning. Fasted. Total, SHBG, free T calculated, LH, FSH, and the labs that find the cause. Then have a provider who can actually read it tell you what it says.

Testosterone therapy, where it’s warranted, is a medical treatment for a diagnosed condition. It requires evaluation, ongoing lab monitoring, and a provider, it carries real risks and side effects, and results vary from person to person. That’s downstream. It’s not the first move and it’s not the point of this page.

The first move is measurement done right. You’ve been carrying a number that was probably wrong before the needle went in. You deserve the real one.

Apex’s lab panels run testosterone the way the guidelines actually describe it: morning, fasted, with SHBG so free testosterone is calculated rather than guessed, plus LH, FSH, thyroid, metabolic, and inflammatory markers so the result comes with a reason attached. A provider reviews all of it with you. Order your panel or book a consult here.

Chris Russell, PA-C is a co-founder of Apex Wellness, a performance medicine clinic specializing in hormone optimization, metabolic health, and longevity medicine. He practices in Georgia with multi-state licensure. Apex Wellness is LegitScript Certified.

This article is for educational purposes only and is not medical advice. It does not establish a provider-patient relationship. Hormone evaluation and any treatment require an individual assessment by a licensed provider, and results vary.

Frequently Asked Questions

How is low testosterone diagnosed?

Low testosterone is diagnosed with a consistent symptom pattern plus at least two total testosterone measurements drawn on separate early mornings while fasting, both below the diagnostic threshold, using an accurate assay. The Endocrine Society’s 2018 guideline requires symptoms plus confirmed low morning testosterone; the AUA’s 2018 guideline uses roughly 300 ng/dL as a reasonable cut-off across two early-morning draws. A complete evaluation also includes SHBG so free testosterone can be calculated, along with LH and FSH to identify whether the cause is testicular or pituitary.

Do you have to fast for a testosterone blood test?

Yes. Eating lowers testosterone measurably. In a 2013 study published in Clinical Endocrinology, Caronia and colleagues found that a standard 75-gram oral glucose load reduced total testosterone by roughly 25 percent on average for up to two hours, and pushed some men with normal levels into the hypogonadal range. A non-fasting draw can produce a falsely low result.

Why does testosterone have to be tested in the morning?

Testosterone follows a circadian rhythm, peaking in the early morning and declining through the day. The diagnostic thresholds used to interpret your result were established on morning values, so an afternoon draw is being compared against a standard it doesn’t match. Both the Endocrine Society and AUA guidelines specify early-morning testing, generally before 10 AM.

What labs should be run besides testosterone?

A complete workup includes SHBG and albumin (so free testosterone can be calculated via the Vermeulen equation rather than measured with an unreliable direct assay), LH and FSH to establish whether the cause is primary or secondary, prolactin, estradiol, a thyroid panel, CBC with hematocrit, a comprehensive metabolic panel, HbA1c and fasting insulin, iron studies, vitamin D, and ApoB. Many of these identify conditions that either mimic low testosterone or cause it.