How to track your testosterone over time
Testosterone is one of the trickier lab tests to read. It changes over the course of a day, it drops after a meal or a few short nights, and two labs can report different numbers for the same blood. A single low result is often not low at all on a repeat draw. That is why the guidelines ask for more than one test before anyone calls it low, and why a history of results, drawn the same way each time, tells you far more than any one of them.
What the test measures
Most testosterone in your blood is bound to proteins. A large share is held tightly by sex hormone-binding globulin (SHBG). Most of the rest is bound loosely to albumin. Only a few percent floats free. A lab report can show any of these:
- Total testosterone: all of it, bound and free. This is the standard test, and the one diagnostic cutoffs are built on.
- Free testosterone: the unbound part only.
- Bioavailable testosterone: the free part plus the part loosely bound to albumin.
- SHBG: the binding protein itself, often ordered alongside so free testosterone can be calculated.
Free testosterone matters most when SHBG is unusual, because then total testosterone can mislead. Obesity, type 2 diabetes, and some medicines lower SHBG, which lowers total testosterone even when the free part is normal. Older age, some anticonvulsants, liver disease, and an overactive thyroid raise SHBG. The Endocrine Society guideline recommends checking free testosterone in exactly those cases, and in men whose total is near the lower limit. It also warns against the “direct” or “analog” free testosterone immunoassays some labs still offer, because they are inaccurate. Free testosterone measured by equilibrium dialysis, or calculated from total testosterone, SHBG, and albumin, is the better number to track. If your report has total testosterone and SHBG but no free value, the free testosterone calculator works it out.
The numbers on the report
Total testosterone in the United States is reported in ng/dL. Much of the rest of the world uses nmol/L. The conversion is:
nmol/L = ng/dL × 0.0347 (or ng/dL = nmol/L × 28.8)
| ng/dL | nmol/L |
|---|---|
| 264 | 9.2 |
| 300 | 10.4 |
| 500 | 17.3 |
| 916 | 31.8 |
Free testosterone comes in smaller units, usually pg/mL or pmol/L (1 pg/mL is about 3.47 pmol/L), and some labs report it in ng/dL. Keep the unit with every free testosterone result. The numbers are not comparable across units, and they are not comparable to total testosterone at all.
Reference ranges
Testosterone ranges have long differed from lab to lab, partly because the tests themselves disagreed. In 2017, a study that recalibrated four large cohorts against the CDC’s reference method set a harmonized range of 264 to 916 ng/dL for healthy, non-obese men aged 19 to 39. Before harmonization, average levels for men in their forties ranged from 501 to 618 ng/dL depending on the cohort. After it, they fell between 471 and 494. Most of that spread was the assay, not the men.
Many labs now print a range close to 264 to 916, but not all do. If your history includes several labs, look at the range printed with each result, not just the number.
The diagnostic cutoff
The AUA guideline treats a total testosterone below 300 ng/dL as a reasonable cutoff in support of a diagnosis of low testosterone. It was published in 2018 and reviewed and confirmed by the AUA in 2024. The Endocrine Society guideline, also from 2018, points to the harmonized lower limit of 264 ng/dL (9.2 nmol/L) for CDC-certified assays, and warns that a local lab’s own lower limit may not identify low testosterone accurately. In July 2026 the Endocrine Society issued a statement on testosterone replacement therapy that refers to a common threshold near 300 ng/dL and restates the core rule.
Both bodies agree on that rule. A diagnosis needs symptoms and consistently low testosterone, confirmed on at least two separate early-morning tests. Symptoms alone are not enough, and neither is one low number.
Testosterone in women
Women’s testosterone is a small fraction of men’s, in the tens of ng/dL rather than the hundreds. Many standard immunoassays were built for the male range and are not accurate that low. The ADLM’s testing guide, following an Endocrine Society position statement, recommends measuring testosterone in women and children with liquid chromatography-tandem mass spectrometry (LC-MS/MS), not immunoassay. If you are a woman tracking testosterone, for example for PCOS, the method matters even more than it does for men. A history that mixes immunoassay and mass spectrometry results may show a step that is only the change of method.
Why the time of the draw matters
Testosterone peaks in the early morning and falls through the day. In a study of 66 men drawn on six separate days, levels at 4 p.m. were 20 to 25 percent lower than at 8 a.m. in men aged 30 to 40. The gap shrank with age, to about 10 percent at 70. Seventeen of the men had at least one afternoon result below 300 ng/dL but were normal on all three morning draws.
Food matters too. In a glucose tolerance study, a 75-gram glucose drink lowered testosterone by about 25 percent on average. The authors advise retesting any low non-fasting result in the fasting state.
That is why the guidelines ask for early-morning, fasting draws. For tracking, it means a result drawn at 3 p.m. after lunch is not comparable to one drawn at 8 a.m. before breakfast. Note the time with every result, and try to keep your draws at the same time of day.
How often it is tested
Testosterone is tested on no fixed schedule unless you are being treated:
- Checking for low testosterone: at least two early-morning fasting draws on separate days. If testosterone is confirmed low, the Endocrine Society recommends LH and FSH to tell whether the problem starts in the testes or in the pituitary.
- After starting testosterone therapy: at 3 to 6 months, according to the Endocrine Society guideline, with a check of symptoms and side effects at 3 to 12 months and then once a year.
- Not on treatment: whenever symptoms or another condition prompt it.
Because the diagnostic draws are often months or years before treatment, the “before” values can be scattered across clinics. Those early results are your baseline. How to get all of your medical records covers tracking them down.
What counts as a real change
Even drawn the same way, testosterone moves from week to week. The European Biological Variation Study, which sampled 38 men once a week for ten weeks, put the within-person variation of testosterone at about 10 percent. Combined with the test’s own imprecision, two results from the same lab can differ by around 30 percent with nothing having changed. On a level of 450 ng/dL, anything from the mid-300s to the high-500s can be ordinary noise.
Add the time of day, food, sleep, and a change of lab, and single results become even harder to read. A few questions help sort it out:
- Were both draws taken the same way? Same time of day, both fasting, same lab. If not, the comparison is weak before you start.
- Is the change bigger than your usual spread? If your morning results have run from 380 to 470, a 340 is not news. A 220 is.
- Did more than one result move? One low value followed by a normal one is the pattern the guidelines’ repeat test is designed to catch.
- Did SHBG move too? A fall in total testosterone with a matching fall in SHBG may leave free testosterone unchanged.
Things that move testosterone
Several things lower testosterone for a while without meaning your baseline has changed:
- Acute illness. The Endocrine Society advises against testing men who are ill or recovering from an illness, because levels can be temporarily low.
- Opioids and glucocorticoids suppress testosterone. Short-term use is another reason the guideline gives not to test.
- Short sleep. In a small study, one week of sleeping about five hours a night lowered daytime testosterone by 10 to 15 percent in young healthy men.
- Weight. Obesity lowers SHBG and total testosterone. The 2026 Endocrine Society statement lists obesity, opioids, and corticosteroids among the reversible causes to rule out first.
- Age. Total testosterone drifts down with age, and SHBG drifts up, so free testosterone falls further than total.
- The assay. Immunoassays and LC-MS/MS can disagree, especially at low levels. The CDC’s Hormone Standardization Program (HoSt) certifies testosterone tests against a reference method, and the 2026 statement points to HoSt-certified assays for consistent results. A report that says “LC-MS/MS” or “MS” in the test name tells you the method.
None of these make the result wrong. They make it a result that needs a note.
Tracking testosterone on treatment
If you take testosterone, the number on the report depends heavily on when it was drawn relative to your last dose. The Endocrine Society guideline gives a timing for each form, including:
| Form | When to draw |
|---|---|
| Enanthate or cypionate injection | Midway between injections |
| Gel | 2 to 8 hours after applying, after at least a week of use |
| Patch | 3 to 12 hours after applying |
| Undecanoate injection | Just before the next injection |
| Pellets | At the end of the dosing interval |
The guideline aims for the mid-normal range on treatment. For an injection, a level drawn the day after a shot and one drawn the day before the next can be hundreds of ng/dL apart. Neither is wrong, but they cannot be compared to each other. Record the date and time of the last dose with each result, and keep the timing the same from draw to draw.
Hematocrit belongs on the same chart. Testosterone raises red blood cell production, and the guideline checks hematocrit at baseline, 3 to 6 months after starting, and then yearly. It advises stopping therapy if hematocrit rises above 54 percent until it comes back down. Seeing both lines together shows whether a dose change moved one without the other. Did the medication actually do anything? covers reading before-and-after results fairly.
What to record with each result
For each testosterone result, keep:
| Field | Why |
|---|---|
| Collection date and time | An afternoon draw can read 20 percent or more lower. |
| Fasting or not | A meal can lower the result. |
| Value and unit | ng/dL or nmol/L for total. pg/mL, pmol/L, or ng/dL for free. |
| Which test | Total, free, or bioavailable, and how free was measured or calculated. |
| Method and lab | Immunoassay or LC-MS/MS. A lab or method change can explain a step. |
| SHBG | Needed to read total testosterone when SHBG is high or low. |
| Last dose | On treatment, the time since the last injection, gel, or patch. |
| Other context | Illness, short sleep, opioids or steroids, a big change in weight. |
Testosterone goes by many names on reports: Testosterone, Total Testosterone, Testosterone, Total, and Testosterone, Total, MS, among others. Free testosterone is a separate test, and so is bioavailable. Merge the total testosterone names into one, keep free and bioavailable on their own lines, and never combine total with free.
Charting it
A few choices make a testosterone chart honest:
- Chart morning, fasting results. If you have an afternoon result, mark it, or leave it out when you read the trend.
- Chart total and free testosterone separately. They are in different units and on very different scales.
- Put SHBG on the same timeline, so a change in total testosterone can be read against it.
- On treatment, add hematocrit, and mark every dose or formulation change.
- Space points by date. Two diagnostic draws a week apart and a follow-up a year later should not look evenly spaced.
How to track your lab results over time has the general method for any test.
Tracking testosterone in HealthViewer
HealthViewer is a lab results tracker that runs on your own computer. It imports testosterone results from lab report PDFs, MyChart downloads, and direct connections to your health system. Results from a direct connection file total testosterone as “Testosterone” and free testosterone as “Free Testosterone”, as separate tests. Total testosterone in nmol/L is converted so it shares one chart with results in ng/dL. Free testosterone in pg/mL and pmol/L is not converted. If your history mixes the two, results in the unit that cannot be converted are left off the chart and listed under it.
Names from PDFs, such as “Testosterone, Total”, “Total Testosterone”, and “Testosterone, Total, MS”, are not combined automatically. In the Test Catalog, check the total testosterone tests and use Merge… to make them one series. Merge total with total only. There is no built-in tag for testosterone, so add your own tag to testosterone, free testosterone, SHBG, LH, and hematocrit, and filter by it in Trends. The overlay view puts testosterone and hematocrit on one chart, and medication markers show where treatment started or stopped. Enter each dose or formulation as its own period in Medications to mark changes too. A journal note on the date of a lab or method change, an afternoon draw, or an illness shows up on the chart when you turn on journal markers.
Everything stays in an encrypted file on your computer, with no account and no cloud service. Getting Started walks through the first import.
This post is general information, not medical advice. Talk to your clinician about your testosterone results and before starting, stopping, or changing any treatment.