How to Test Testosterone Levels: Step-by-Step

To test testosterone levels, get a clinical blood draw between 7-10 AM after a fast, and make sure the lab uses a CDC-certified assay. A single test is a snapshot; due to natural biological variation of 10-15%, a truly low result usually needs a second morning draw for confirmation. At-home saliva or blood-spot kits offer convenience but lack the standardized accuracy of a clinical lab for diagnosis.

That 10-15% biological variation is the whole game. It means two tests on the same healthy person can differ by up to 30% and both be correct. Chasing a number from a single afternoon draw, or reacting to a result that’s just shy of an arbitrary line on a chart, is how men end up confused or on unnecessary therapy. Understanding how to test testosterone levels properly is therefore critical.

What follows is the mechanics: how a blood draw works, where home tests fit, how to read the report without panicking, and the one sample-handling mistake that can inflate your result by 50% before it even reaches the lab, all central to the how to test testosterone levels question.

Key Takeaways

• Schedule your blood draw between 7 and 10 AM, fasted, after a normal night’s sleep. Testosterone has a strong diurnal rhythm, and afternoon levels can be lower.
• Verify the lab uses a CDC-certified assay. The CDC’s standardization program certifies that specific commercial test methods meet a strict ±6.4% accuracy bias against their gold-standard method.
• View any single result as a data point, not a verdict. Healthy biological variation means two tests can differ by up to 30%. A genuinely low level may require a second morning test for confirmation, as recommended by some authorities, while others note a repeat may not be needed if a result within the 20–30% range would not impact the choice of treatment.
• Separate serum from red blood cells within 6 hours. If the sample sits at room temperature longer, enzymes in the red cells can convert other hormones into testosterone, artificially inflating your result.
• For a basic check, a total testosterone test is standard. If that result is borderline low or you have conditions like obesity, a calculated bioavailable testosterone (cBAT) test that includes SHBG is the necessary next step.

The Three Types of Testosterone Tests

You are not ordering one test. You are choosing which fraction of your circulating testosterone to measure, and that choice changes what the number means.

A total testosterone test is the default and the most common. It counts every molecule of testosterone in your blood, both the fraction bound tightly to proteins like sex-hormone-binding globulin (SHBG) and the fraction that is free or loosely bound. This is the number most lab ranges and clinical guidelines are built on. It is a solid starting point.

A free testosterone test measures only the unbound, biologically active hormone, the part your tissues can actually use immediately. This test is less common and more technically challenging to standardize. It becomes crucial when SHBG levels are very high or very low, which can happen with obesity, thyroid issues, or liver disease. In those cases, total testosterone can look normal while free testosterone is actually low.

A calculated bioavailable testosterone (cBAT) test is needed when total testosterone is low but the clinical picture is unclear. It uses your total testosterone and SHBG levels to estimate the fraction that is not tightly bound to SHBG, which includes both free testosterone and testosterone loosely attached to albumin.

A bioavailable testosterone test (or the more common calculated version, cBAT) measures the free testosterone plus the portion loosely bound to albumin. Think of it as the “available for duty” pool. The British Columbia clinical protocol explicitly states this test is needed to clarify a diagnosis when total testosterone is low but other questions remain. For most initial screenings, total testosterone is sufficient. The red flag is when that number is low; that is when you need the extra data point of SHBG to calculate cBAT, a key nuance in how to test testosterone levels.

The Step-by-Step Clinical Blood Test

This is the gold standard. The procedure is simple, but the details around timing and handling are where accuracy is won or lost.

1. Schedule the Draw for the Right Time

Testosterone peaks in the early morning and declines through the day. The Province of British Columbia guidelines specify collection before 10:00 AM, or within three hours of waking. MedlinePlus narrows it to between 7 and 10 AM. For shift workers, this protocol breaks down, your “morning” is after your main sleep period, but the diurnal rhythm is sleep-driven, not purely circadian. Interpretation in this case is complicated and often warrants specialist input.

2. Prepare Your Body

Be clinically well. Do not test during an acute illness, infection, or hospitalization, as these can temporarily suppress levels. The guidelines are split on fasting: the BC protocol says “preferably in a fasting state,” while MedlinePlus notes you may need to fast for several hours in certain cases. The conservative approach is to fast, as recommended by some sources. Get a normal night’s sleep beforehand. Testing during jet lag or right after a shift change adds noise.

3. The Blood Draw and Critical Sample Handling

A phlebotomist will collect a vial of blood from your arm. The science happens after you leave. The sample must be centrifuged to separate the liquid serum from the red blood cells. According to the CDC’s laboratory method, this separation must happen within 6 hours at room temperature or 24 hours at 4°C to avoid enzymatic conversion. Wait longer, and enzymes in the red cells start converting a precursor hormone (androstenedione) into testosterone. This can artificially inflate your measured level by up to 50% after 24 hours at room temperature. Once separated, serum for long-term storage is kept at -70°C.

4. Lab Analysis and the Importance of Assay Choice

Your serum is analyzed. This is the most important variable you rarely see. Many labs use automated immunoassays, which can vary in accuracy. The CDC runs a Clinical Standardization Program that certifies specific commercial assays and reagent lots that meet a performance criterion of ±6.4% mean bias compared to their reference method. Certification is valid for one quarter. The higher-accuracy method is liquid chromatography tandem mass spectrometry (LC-MS/MS). When you order a test, you are trusting the lab’s choice of tool. Some direct-to-consumer services, like Function Health, explicitly use CDC-certified or LC-MS/MS methods.

Where this goes sideways: Drawing blood in the afternoon after a poor night’s sleep and treating that single number as an absolute baseline. The result will be low, but it reflects circumstance, not necessarily your underlying hormonal state.

Step	Key Action	Why It Matters	Risk If Skipped
Scheduling	Book between 7-10 AM	Captures peak diurnal level	Afternoon levels can be lower, suggesting deficiency where none exists.
Preparation	Fast, sleep well, be healthy	Minimizes temporary suppressors	Acute illness or poor sleep can artificially lower results, leading to misdiagnosis.
Sample Handling	Separate serum from RBCs within 6 hours	Prevents enzymatic conversion	Testosterone levels can be artificially inflated by up to 50% after 24 hours at room temperature, hiding a true low.
Lab Analysis	Use a CDC-certified or LC-MS/MS assay	Ensures analytical accuracy	Unstandardized assays can have significant bias, making trends or comparisons unreliable.

Understanding Your Test Results and “Normal” Ranges

The number comes back. Now you have to decide what it means, which is harder than it should be.

First, look at the reference range printed on your lab report. These ranges are established by the lab using data from their specific testing population and method. They often blend wide age groups. A man in his 50s seeing a range built from 20- to 70-year-olds is comparing himself to a statistical blur. More nuanced guidelines, like those from British Columbia, suggest age-adjusted cut-offs: approximately 8 nmol/L (230 ng/dL) for men under 30 and 6 nmol/L (170 ng/dL) for men over 50. Other authorities, like a UK joint position statement, propose a single action threshold of <8 nmol/L for all symptomatic adult males, while guidelines like those from British Columbia suggest age-adjusted cut-offs of approximately 8 nmol/L for men under 30 and 6 nmol/L for men over 50.

Second, understand biological variation. The BC guidelines note a healthy adult male’s levels can fluctuate by 10-15% due to normal daily and weekly rhythms. This means two successive measurements can differ by up to 30% without anything being wrong. This is why a single low test is not a diagnosis.

I tested myself on a few different mornings over several weeks. The swing between draws was bigger than I expected, sometimes over 100 ng/dL. A single draw is a snapshot, not a verdict. Now I only compare fasted morning draws, and I repeat any odd result before reacting.

Common mistake: Someone in their mid-50s pulls up an online chart built around 20-somethings and assumes something is broken. Most lab ranges blend wide age groups, so the comparison is mostly noise. Look at age-stratified ranges if available, and focus on symptoms.

When is a result truly “low”?

Late-onset hypogonadism (LOH) is defined as symptoms in a man over 65 with a serum testosterone concentration less than 6-8 nmol/L. For younger symptomatic men, most guidelines point to levels consistently below approximately 8 nmol/L as likely deficient. The critical link is symptoms: low libido, unexplained fatigue, loss of muscle mass, depressed mood. Testing asymptomatic men is not recommended.

If your total testosterone is low and you have symptoms, the next step often isn’t treatment, it’s more precise testing. A calculated bioavailable testosterone (cBAT) test, which requires a SHBG measurement, is the necessary follow-up. This tells you how much hormone is actually available to your tissues.

At-Home Testing Kits: Convenience vs. Accuracy

You can now order a test online, prick your finger or spit in a tube, mail it in, and get results digitally. The appeal is obvious: privacy, no doctor’s visit, and often a lower upfront cost. But you trade control over the pre-analytical variables and, sometimes, analytical rigor.

Most home kits are blood-spot tests. You lance your finger, blot blood onto a special card, and mail it to a partner lab. Others use saliva collection. The major companies in this space include LetsGetChecked, Everlywell, and private lab direct-access services like Labcorp OnDemand or Quest Direct.

The convenience is real. The trade-off is in the pre-analytical phase. You are responsible for the timing, your fasting state, and the sample stability during shipping. A sample mailed in a hot mailbox for two days is not the same as a properly handled clinical sample. Many home kits use the same decent lab assays as clinical tests, but the sample type (dried blood spot or saliva) itself has limitations.

Saliva testing for testosterone is particularly contentious. While it measures free testosterone, its correlation with serum free testosterone is not strong enough for diagnostic purposes. Studies show higher variability and a lack of standardized reference ranges. For tracking broad trends in a healthy person, it might offer a signal. For diagnosing hypogonadism, it is not reliable.

I recommend against at-home saliva kits for diagnosis. The data on their reliability compared to blood is thin, and you can’t afford to be wrong on something that might lead to a treatment decision.

So, are at-home tests accurate? They can be precise for the sample type they use. But accuracy in the medical sense, giving you a number that a doctor would use to confidently diagnose or rule out a condition, is lower than a controlled venous blood draw. They are excellent tools for curiosity, initial screening, or monitoring a known level if you always use the same kit. They are poor tools for a definitive diagnosis.

Why You Might Need a Test (And Why You Might Not)

Testing should be driven by symptoms and clinical suspicion, not curiosity or a “wellness” check.

Get tested if you are a man experiencing:

• Persistent, unexplained fatigue and low energy
• Markedly reduced libido or erectile dysfunction (though ED alone is not a sufficient reason)
• Loss of muscle mass and strength despite exercise
• Increased body fat, particularly around the abdomen
• Depressed mood, irritability, or poor concentration
• Hot flashes or low bone mineral density (in older men)

Do not get tested if:

• You are asymptomatic. “Wellness” screening is not recommended.
• You have isolated erectile dysfunction without other symptoms.
• You are a woman investigating low libido (testosterone testing is not useful for this in women).
• You are acutely ill or hospitalized.

The process of diagnosing low testosterone requires both consistent symptoms and corroborating lab work. A doctor will also test other hormones like LH and FSH to determine if the cause is primary (testicular failure) or secondary (a pituitary/hypothalamic issue). They will also rule out other causes like sleep apnea, thyroid disorders, or depression.

How to Talk to Your Doctor and Next Steps

You have your results. Now you need a clinical conversation.

First, get the actual lab report, not just a “normal” or “low” from a portal. Look at the numbers, the units (ng/dL or nmol/L), and the reference range. If you used a direct-to-consumer testing service, you will bring these results to your doctor.

When you meet, frame it around symptoms, not just a number. Say, “I’ve been struggling with severe fatigue and low drive for six months. I got this test to explore it, and my level came back at X. Can we discuss what this means in the context of my symptoms?” This is collaborative, not confrontational.

Be prepared for your doctor to want a repeat test. Given the 10-15% biological variation, a single low result is not conclusive. Most responsible guidelines, including a major UK position statement, recommend confirmation with a second morning draw. The BC guidelines offer a nuance: if a difference of 20-30% wouldn’t change the treatment decision (e.g., symptoms are severe and the level is very low), a repeat may be skipped. But in borderline cases, a second test is standard.

If deficiency is confirmed, the discussion turns to investigation of the cause and treatment options, which may include lifestyle changes or replacement therapy. Remember, starting testosterone therapy suppresses your natural production for as long as 18 months, so the diagnostic certainty needs to be high.

Testing for women/transgender individuals

Testosterone testing for women and transgender individuals follows the same core procedures for sample collection and handling, but the clinical context and interpretation are different. The blood sample must be collected in the morning, preferably before 10:00 AM, and when the individual is clinically well, not during an acute illness. The serum must be separated from red cells promptly to avoid artificially inflated results. For individuals undergoing prolonged androgen therapy, it is known that such treatment can suppress the body’s own hormone production for an extended period. It is not recommended to test asymptomatic individuals, and testing is not considered useful for investigating low libido in women. The result must be interpreted by a healthcare provider familiar with the specific clinical situation and relevant reference ranges.

Testing for children/puberty issues

When testing testosterone levels in children or adolescents for concerns about puberty, the focus is on precise laboratory methods and proper sample timing. The testing should occur when the child’s sleep-wake pattern is stable and they are clinically well, not during an acute illness. The blood draw should be done in the morning. Accurate measurement is critical, and laboratories may use highly sensitive methods to detect very low levels. The serum sample requires careful handling; it must be separated from red cells within a specified time to prevent conversion of other hormones into testosterone, which would give a falsely high reading. Interpreting the results requires a pediatric specialist who can compare the value to age and puberty-stage-appropriate reference ranges.

Frequently Asked Questions

How do I check my testosterone level at home?

You can order a kit online from companies like LetsGetChecked or Everlywell. Most involve a finger-prick blood sample you mail to a lab. Saliva kits also exist. While convenient for a general look, these home testosterone tests lack the rigorous control of a clinical venous draw and are not recommended for definitive diagnosis. Use them for initial screening or trend monitoring, not for medical decisions.

What is a normal testosterone level by age?

Lab reference ranges vary. Lab reference ranges for total testosterone vary. Some clinical protocols use age-adjusted cut-offs, like approximately 8 nmol/L (230 ng/dL) for men under 30 and 6 nmol/L (170 ng/dL) for men over 50. “Normal” is a wide band, and being in the lower part of the range without symptoms is not a disease.

How do I know if my testosterone is OK?

Match your numbers to your symptoms. If your level is well within your lab’s reference range and you have no symptoms of deficiency (persistent fatigue, low libido, mood changes), it is likely OK. If you have clear symptoms and your level is at or below the bottom of the range, it warrants a discussion with a doctor. A single test is just one piece of information.

Is 9.5 nmol/L testosterone low?

It depends on your age and symptoms. For a young man under 30, 9.5 nmol/L is below the approximate 8 nmol/L action threshold suggested by some guidelines and would be considered low if accompanied by symptoms. For a man over 50, it is above the 6 nmol/L threshold and may not be considered deficient. The context of symptoms and a repeat test are essential for interpretation.

Before You Go

Get the first test right. That means a morning, fasted blood draw at a reputable lab. It is the only way to get a clean baseline number you can actually trust.

If that first result is low, do not panic. Schedule a second morning draw before you do anything else. Biological variation is real, and you need two points to establish a trend. This is the step that prevents men from treating a temporary dip or a lab error.

Finally, look beyond the testosterone number alone. A comprehensive hormone panel shows you why testosterone might be low, through markers like LH and FSH, and checks related nutrients like vitamin D. Correcting a simple, common deficiency is often where the real gains are made, long before more complex interventions are needed.