Testosterone Levels by Age: Normal Ranges and What They Mean

For healthy, non-obese men aged 19 to 39, the harmonized reference range for total testosterone is 264 to 916 ng/dL, with a median of about 531 ng/dL. That range comes from a 2017 study that standardized measurements across four large cohorts against the CDC's reference method, and it is the basis of the Endocrine Society's diagnostic threshold. Most clinical laboratories still print a broader adult range, typically around 300 to 1,000 ng/dL. Levels peak in the late teens and twenties and fall by roughly 1% a year from the thirties onward.

The first thing to know about testosterone by age is that there is no single agreed table of decade averages. Different labs use different assays, different studies sampled different men, and most of the neat charts that circulate online cite no source at all. What is well established is the shape of the curve, a well-derived young-adult range to anchor it, and the average rate of decline. What is not established is that a 48-year-old should have exactly some number.

This guide works through the whole arc: how testosterone is measured and why timing matters, what free testosterone adds, what happens from infancy through the seventies, why levels fall, why the charts disagree, what counts as low, and how any of it should feed a decision about treatment. The short version is our normal testosterone levels by age explainer; the treatment side is covered in the complete guide to TRT and in is TRT safe?. None of it is medical advice, and your own lab's reference range is the one that applies to your result.

Testosterone is the main male sex hormone, made mostly in the testes under instruction from the pituitary gland. It drives puberty, then maintains libido, erections, sperm production, muscle and bone mass, red blood cell production, fat distribution, mood and energy through adult life. Because it touches so many systems, a level that is genuinely too low produces a scattered set of symptoms, and a number that is normal for your age is compatible with feeling entirely well.

Production is a feedback loop. The hypothalamus releases GnRH in pulses; the pituitary responds with luteinizing hormone (LH) and follicle-stimulating hormone (FSH); LH tells the Leydig cells of the testes to make testosterone, and FSH with local testosterone drives sperm production. When circulating testosterone is adequate the brain eases off. A problem can therefore sit in the testes (primary hypogonadism, with high LH) or in the brain's signalling (secondary hypogonadism, with low or inappropriately normal LH), and the distinction changes both the search for a cause and the treatment.

Most testosterone in blood is bound to proteins, chiefly sex hormone-binding globulin (SHBG) and more loosely albumin. Only the unbound fraction, plus the loosely albumin-bound portion, is available to tissues. That detail becomes important with age, as the section on free testosterone explains. For a primer on the hormone and its replacement, see what is TRT?.

The best-derived modern reference is the harmonized range published by Travison and colleagues in the Journal of Clinical Endocrinology and Metabolism in 2017: 264 to 916 ng/dL for healthy non-obese men aged 19 to 39, representing the 2.5th to 97.5th percentiles, with a median of 531 ng/dL, a 5th percentile of 303 and a 95th of 852. Common laboratory ranges run roughly 300 to 1,000 ng/dL. The Endocrine Society adopted 264 ng/dL as its lower limit; the American Urological Association uses 300 ng/dL.

The harmonized study matters because it solved a measurement problem. Before it, every cohort used its own assay and the numbers were not comparable; the investigators re-standardized samples from the Framingham Heart Study, the European Male Aging Study, the Osteoporotic Fractures in Men study and the Boston Area Community Health survey against the CDC's liquid chromatography-mass spectrometry reference, which is why its range is now the anchor. Two caveats travel with it: it describes men under 40 who were not obese, and it is a statistical band, not a target. A healthy man can sit anywhere in it. For readers outside the US, 1 nmol/L is 28.84 ng/dL, so 264 to 916 ng/dL is roughly 9.2 to 31.8 nmol/L.

Testosterone is measured from a blood sample as total testosterone, in ng/dL in the US and nmol/L elsewhere, drawn in the early morning because levels peak between about 7 and 10 a.m. and fall through the day. Guidelines require two separate morning samples before anyone is called low, because a single reading can be depressed by the time of day, a poor night's sleep, acute illness, recent heavy exercise or fasting state.

The circadian swing is large enough to move a man across a diagnostic line. In younger men the afternoon trough can be 20 to 30% below the morning peak; the rhythm flattens with age but does not disappear. Reference ranges were built from morning samples, so an afternoon value is being compared with the wrong yardstick. The Endocrine Society guideline specifies fasting morning samples and repeat confirmation, and a clinic that diagnoses from one afternoon draw is cutting a corner that matters.

The assay matters too. Immunoassays, which most routine labs use, are less accurate at low concentrations than mass spectrometry and vary between manufacturers, which is one reason ranges differ between labs. Free testosterone adds another layer: direct immunoassays for free testosterone are unreliable, so guidelines prefer equilibrium dialysis or a calculated value from total testosterone, SHBG and albumin. Our guide to TRT blood work covers what each test is for; how to get TRT online explains how telehealth clinics arrange proper morning draws.

Free testosterone is the 1 to 3% of total testosterone not bound to any protein, and it is the fraction tissues can use. SHBG, the main binding protein, rises steadily with age, so an older man's free testosterone can fall further than his total suggests. That is why guidelines recommend measuring or calculating free testosterone when total is borderline, when SHBG is likely abnormal, or when symptoms and the total number disagree.

Several things move SHBG. Ageing, liver disease, hyperthyroidism, some anticonvulsants and HIV raise it; obesity, insulin resistance, type 2 diabetes, hypothyroidism and nephrotic syndrome lower it. A man with obesity may therefore have a total testosterone that looks low while his free fraction is closer to normal, and a lean 70-year-old may have a reassuring total and a genuinely low free level. In the Baltimore Longitudinal Study of Aging, total testosterone fell by about 1% a year while the free testosterone index fell by about 2%, almost entirely because SHBG was rising (Harman et al., 2001).

Practically: if your total is between roughly 200 and 400 ng/dL and you have symptoms, ask for SHBG and a calculated free testosterone. If your report already lists a free value from a direct immunoassay, treat it cautiously. Our free versus total testosterone guide works through the arithmetic.

The curve has six phases: a brief surge in the first months of life, low and stable levels through childhood, a steep rise through puberty, a peak and plateau from the late teens through the twenties, a slow decline beginning in the thirties, and a continued decline through older age in which free testosterone falls faster than total. The shape is well established; the exact numbers at each age are not.

The table is deliberately directional. A chart that assigns a single average to each decade would look more authoritative and be less true, because the published cohort averages differ by assay and population by tens of ng/dL. What the data agree on is the sequence and the slope, and those are what let you place your own number in context. The sections that follow take each phase in turn.

Newborn boys have a transient rise in testosterone that peaks at around one to three months of age and fades by about six months, a phase endocrinologists call mini-puberty. From then until puberty, levels are low and stable, often below the detection limit of routine assays. Adult reference ranges have no meaning in this period.

The prenatal and infant surges matter developmentally; they shape the genital tract and may prime later reproductive function. Clinically, the infant window is sometimes used by paediatric endocrinologists to assess the reproductive axis in boys with undescended testes or ambiguous findings, because it is the only time before puberty that the axis is active. After it closes, the hypothalamus and pituitary fall quiet for a decade.

Concerns in childhood, signs of puberty appearing before about age 9 or failing to appear by about 14, belong with a paediatrician, not with an adult testosterone chart. Everything that follows applies to men who have completed puberty.

Puberty begins when the hypothalamus resumes pulsatile GnRH release, usually between 9 and 14, and testosterone climbs from childhood levels into the adult range over three to five years. It drives testicular and penile growth, the growth spurt, voice change, body and facial hair, muscle development, the start of sperm production and the emergence of adult libido and mood. The timing varies widely and the sequence matters more than the age.

Paediatricians stage puberty by physical development, testicular volume and growth velocity rather than by comparing a boy's testosterone to adult norms, because levels during puberty are a moving target. Delayed puberty, no testicular enlargement by 14, is the commonest reason a teenage boy has his testosterone measured, and most cases are constitutional delay rather than disease. Precocious puberty, before 9, is investigated for the opposite reason.

By the late teens most young men have reached the adult range and are approaching their lifetime peak. Testosterone at 17 is already adult testosterone, which is one reason anabolic steroid use in late adolescence, when the axis has only just matured, is especially damaging to it. Our guide to what causes low testosterone covers the developmental causes that occasionally surface at this stage.

In the late teens and twenties. Population studies put the highest average total testosterone around ages 19 to 30, after which the plateau holds into the early thirties for most men before the gradual decline begins. In the harmonized cohort of 19- to 39-year-olds the median was 531 ng/dL; the younger end of that window sits a little higher than the older end.

Peak does not mean a spike. Testosterone reaches a plateau and holds it, with day-to-day and season-to-season variation around a stable mean, for roughly a decade. The timing of the subsequent decline differs between men by years, which is one reason two 35-year-olds with identical lifestyles can have quite different levels.

The peak is also the implicit comparator in every "low for my age" worry. A man of 55 comparing himself with the levels he had at 25 is comparing himself with the top of the curve, and the gap he sees is mostly the curve, not a disease. Whether anything in that gap needs treating is the question the rest of this guide addresses.

In the twenties, total testosterone sits at or near its lifetime plateau; the harmonized range of 264 to 916 ng/dL, median 531, was built largely from men of this age, and most healthy men fall in the upper half of any laboratory range. Low testosterone in this decade is uncommon and, when it is real, usually has a specific cause rather than age.

The causes worth finding in a young man with two low morning readings and symptoms include a pituitary problem such as a prolactinoma, Klinefelter syndrome and other primary testicular conditions, prior anabolic steroid use, opioid medication, severe obesity, untreated sleep apnoea, and chronic illness. The workup, LH and FSH to localize the problem, prolactin, iron studies, a medication and supplement history, is short and often decisive, and several of the findings are reversible without lifelong replacement.

For the large majority of men in their twenties, testosterone is not a concern, and the levers that keep it where it should be are ordinary: sleep, a healthy weight, resistance training, moderate alcohol and no anabolic steroids. The "optimization" market that encourages healthy young men to test repeatedly and chase the top of the range is selling a number, not health. Our signs of low testosterone guide is the place to start if symptoms are real.

The thirties are still inside the harmonized reference population, so 264 to 916 ng/dL remains the working range, but this is the decade in which the slow decline usually starts: about 1% a year on average in total testosterone, almost imperceptible year to year and meaningful only when it accumulates over decades. Healthy men in their thirties remain well within the normal range and rarely have symptoms attributable to age alone.

What changes in this decade is often not the hormone but the life around it: longer working hours, less sleep, weight gain, more alcohol and less training. Each of those lowers testosterone independently of age, and a 36-year-old with a low morning level who has gained 30 pounds and sleeps six hours is more likely to be looking at those factors than at his birth certificate. The decline attributable to age at this stage is small; the decline attributable to lifestyle can be large.

The practical message for the thirties is to attend to the modifiable factors first and to take persistent symptoms seriously rather than assuming they are age. Our guide to natural ways to support testosterone covers what the evidence supports, and what the supplement aisle does not.

By the forties the cumulative decline is typically 10 to 15% from the young-adult peak in total testosterone and more in free testosterone as SHBG rises, yet most healthy men remain inside the laboratory reference range. This is the decade in which men most often first test their testosterone, usually prompted by fatigue, lower libido or difficulty keeping weight off, and the decade in which the question "age or something else?" is hardest to answer from the number alone.

Three things make interpretation in the forties genuinely tricky. The symptoms men bring are nonspecific and overlap with sleep deprivation, stress, depression, thyroid disease and simple deconditioning. The total testosterone can look adequate while the free fraction, the one tissues feel, has fallen more. And lifestyle contributors, especially central weight gain, are at their most prevalent. A man of 45 with a total of 340 ng/dL and symptoms should have his SHBG and calculated free testosterone measured, his sleep and weight examined, and the test repeated, before anyone decides what the number means.

The two errors to avoid are symmetric: dismissing genuine symptoms as middle age, and medicating a number that is normal for a man's age because a marketing page said it was low. The first misses treatable hypogonadism; the second exposes a man to the risks described in is TRT safe? for no benefit.

In the fifties the average total testosterone is clearly below the young-adult median, a larger share of men fall below 300 ng/dL, and free testosterone has usually declined further than total. In the European Male Aging Study, which measured 3,369 men aged 40 to 79, the prevalence of late-onset hypogonadism defined by both low hormones and sexual symptoms was 2.1% overall and rose with age. "Lower than at 25" is expected in this decade; "clinically deficient" is still the minority.

The EMAS criteria are worth knowing because they are the most rigorous attempt to separate ageing from disease in this age group: three sexual symptoms (reduced morning erections, reduced sexual thoughts, erectile dysfunction) together with a total testosterone below 317 ng/dL (11 nmol/L) and a free testosterone below 64 pg/mL (220 pmol/L). Men who met them were a small fraction of those with a low number alone. The study is Wu et al., New England Journal of Medicine, 2010.

Interpretation in the fifties therefore leans harder on symptoms, free testosterone and the search for other causes, because fatigue, low mood and weight gain at this age have many parents. A man who meets the full picture, two low morning readings, symptoms and no reversible explanation, is a legitimate candidate for a treatment conversation; a man with a number and nothing else is not. The low testosterone symptoms guide covers how to tell the two apart.

In the sixties and seventies, average total testosterone continues to fall, a substantial minority of men have values below the young-adult range, and free testosterone is lower still because SHBG keeps rising. In EMAS the prevalence of symptomatic late-onset hypogonadism reached about 5% in men aged 70 to 79. Most older men with a "low" number by a young man's standard are asymptomatic, and guidelines and the FDA label agree that a lower level with age, without symptoms and a diagnosis, is not a reason to treat.

Older men are also the group in which other causes of a low reading are most common: chronic illness, medications, obesity, and acute illness at the time of testing, which can lower testosterone sharply and temporarily. They are the group for which free testosterone is most informative, and for which the consequences of true deficiency, anaemia, bone loss and loss of muscle, are most serious. Both over- and under-treatment carry real costs here.

The balanced approach: expect lower levels; measure properly, in the morning, twice, with SHBG; take a confirmed, symptomatic deficiency seriously; and, if treatment is considered, insist on the closer monitoring that age warrants, as is TRT safe? describes for older men. TRAVERSE, the large safety trial, enrolled men up to 80 with a mean age of 63, so its reassurance applies to this group when the diagnosis is real.

Because several slow changes add up: the Leydig cells of the testes become fewer and less responsive, the pituitary's LH pulses become weaker and less regular, SHBG rises and binds a larger share of what is made, and the conditions that accumulate with age, weight gain, poorer sleep, type 2 diabetes, chronic illness and medications, each lower testosterone independently. The decline is gradual because no single switch is thrown.

• Testicular ageing. Reduced Leydig cell number and function and a blunted response to LH.
• Central ageing. Less robust GnRH and LH pulsatility, so the instruction to the testes weakens.
• Rising SHBG. More of the hormone is bound and less is available, which is why free testosterone falls about twice as fast as total.
• Accumulated health changes. Visceral fat converts testosterone to estradiol and suppresses LH; sleep apnoea, diabetes, opioids and glucocorticoids all lower levels.

The encouraging corollary is that a meaningful part of "age-related" decline is health-related decline wearing age as a disguise. Longitudinal data show that men who stay lean, sleep well and avoid chronic illness lose testosterone more slowly than the population average, and that weight loss in men with obesity raises it. Age sets the slope; health decides where on it you sit. Our guide to what causes low testosterone takes each driver in turn.

On average, total testosterone falls by about 1% a year from the thirties onward and free testosterone by about 2% a year, figures that come from the Baltimore Longitudinal Study of Aging, which followed 890 men with repeated measurements over decades. The range around those averages is wide, and the rate for an individual depends more on weight, sleep, health and medications than on his birth year.

One percent a year is small in any single year and large over thirty: a man whose total testosterone was 600 ng/dL at 30 and who follows the average slope is near 450 at 60. The free testosterone index falls faster because SHBG rises at roughly the same time, which is why symptoms sometimes appear in men whose total number still looks only modestly reduced. The BLSA paper is Harman et al., JCEM 2001; its finding has been replicated in European cohorts with similar slopes.

The individual variation is the practical point. In the same cohorts some men showed no decline at all into their sixties, and the men who declined fastest were disproportionately those who gained weight or developed chronic illness. The 1% figure describes a population; your own trajectory is partly in your hands.

The decline itself is normal. It becomes a medical problem only when the level falls low enough, measured properly and twice, to produce symptoms that fit and that are not better explained by something else. Most men never cross that line; their levels ease downward and stay adequate. A smaller group develop what guidelines call late-onset hypogonadism, which requires both the low number and the clinical picture.

The distinction protects against two errors. One is turning normal ageing into a disease and prescribing testosterone to men who are simply older than they were, which exposes them to monitored risks for uncertain benefit and is what the FDA's label specifically warns against. The other is dismissing a real, treatable deficiency as "just age" and leaving a man anaemic, osteopenic and miserable when a diagnosis was available.

The line between them is drawn by process rather than by a number: morning testing, repeated; SHBG and free testosterone where the total is borderline; a symptom inventory; a search for reversible causes; and a clinician's judgement. The complete guide to TRT describes how that process leads, or does not lead, to treatment.

"Male menopause" and "andropause" are popular names for age-related testosterone decline and its symptoms. The phenomenon is real; the analogy is poor. Female menopause is a defined event in which ovarian hormone production falls steeply over a few years and happens to every woman. Male testosterone falls slowly over decades, at rates that vary enormously, and many men never develop a deficiency at all. The medical term for the minority who do is late-onset hypogonadism.

The EMAS investigators built their criteria precisely to find out how many middle-aged and older men have a genuine syndrome rather than a lower number, and the answer, around 2% overall, is the best evidence that andropause is not a universal life stage. Many men who attribute fatigue, low mood and low libido to "male menopause" turn out on evaluation to have normal-for-age levels and a different, often treatable, explanation: sleep apnoea, depression, thyroid disease, a medication or simple overwork.

The right response to the worry is therefore a proper evaluation rather than either dismissal or a prescription. If the evaluation finds a real deficiency, it is treatable. If it finds something else, that something else usually has its own solution. Our signs of low testosterone guide helps sort the symptoms.

Because they are built from different things. Laboratories use different assays that read the same sample differently. Studies sample different populations, lean young men in one, a cross-section of a community in another. "Normal" is defined by different percentile cut-offs. And a large share of the decade-by-decade tables online cite no study at all and were assembled to sell testing or treatment. The harmonized young-adult range exists because researchers recognised this problem and fixed it for one age group.

• Assay. Immunoassays differ by manufacturer and read low concentrations poorly; mass spectrometry is the reference and gives lower values in some ranges.
• Population. Obesity lowers testosterone, so a range from a general population sits below one from non-obese men; age and health mix in the same way.
• Statistics. A 2.5th to 97.5th percentile band is wider than a 5th to 95th band from the same data.
• Provenance. Many charts are copied from other charts; the original source, if any, is lost.

This is why the present guide anchors on one well-derived range and declines to print a false-precision table of decade averages. The empowering version of the same point: be sceptical of any chart that tells you a 45-year-old should have a specific number, read your result against the range printed on your own report, and let a clinician interpret the borderline cases.

Two reasons. Each laboratory uses a particular assay, and assays do not agree perfectly with one another, especially at the low end. And each laboratory builds or adopts its reference interval from a particular population, so the band it prints reflects who was sampled. Both are legitimate, both are why the same blood can produce different numbers and be judged against different cut-offs in two labs, and both are why the range on your own report is the one that counts.

The variation is large enough to matter at the margin. A total testosterone of 290 ng/dL is below range at a lab whose lower limit is 300 and inside it at a lab that uses the harmonized 264. Neither lab is wrong; the diagnosis simply cannot rest on where the line falls. It rests on repetition, on symptoms and on free testosterone, which is the guideline position and the clinically sensible one.

For a man tracking his own levels over time, the practical rule is to use the same laboratory and the same time of day each time, so that changes reflect him rather than the method. For a man comparing his number with a friend's, the rule is not to bother.

Strongly, and in both directions. Excess visceral fat lowers testosterone through several routes: the aromatase enzyme in fat tissue converts testosterone to estradiol, which suppresses LH; obesity lowers SHBG; and it travels with insulin resistance, type 2 diabetes and sleep apnoea, each of which lowers testosterone on its own. Men with obesity have, on average, markedly lower total testosterone than lean men of the same age, which is why the harmonized reference range was derived in non-obese men.

The loop runs backwards too. Low testosterone makes it harder to build muscle and easier to accumulate fat, which lowers testosterone further. The clinically useful fact is that the loop can be reversed: in longitudinal data, including EMAS, men who lost weight saw their testosterone rise, and the rise scaled with the weight lost. For a man with obesity and a borderline level, meaningful weight loss is both a treatment for the low number and a test of whether it needed anything else.

This is also why a good clinic weighs a man before it prescribes to him, and why GLP-1 medication has become part of the low-testosterone conversation for men whose weight is the driver. Our weight-loss hub covers that route in detail; the TRT guide to supporting testosterone naturally covers the rest.

Sleep deprivation and untreated sleep apnoea, opioid medications, glucocorticoids such as prednisone, prior anabolic steroid use, chronic illness including type 2 diabetes and kidney or liver disease, heavy alcohol use, chronic stress, pituitary disease including prolactin-secreting tumours, and acute illness or surgery at the time of the test. Several of these are reversible, and all of them can make a man look "low for his age" when age has nothing to do with it.

• Sleep. Most daily testosterone is produced during sleep; restricting sleep to five hours for a week lowered testosterone by 10 to 15% in young men in a small controlled study, and apnoea does the same chronically.
• Opioids. Long-term opioid use suppresses the axis profoundly and is one of the commonest unrecognised causes in middle-aged men.
• Prior steroids. Anabolic steroid use shuts down the body's own production, sometimes for years after stopping.
• Chronic disease. Diabetes, obesity-related and otherwise, and any serious systemic illness.
• Acute illness. A level drawn during a flu or after surgery can be half the man's usual, which is why testing waits until he is well.

The reason this list belongs in a guide about age is that each item is a competing explanation that a thorough clinician excludes before attributing a low level to the calendar. A man who finds his opioid prescription or his apnoea on this list may have found the whole answer. Our causes guide goes through each.

A morning total testosterone below about 300 ng/dL (the American Urological Association's threshold) or below 264 ng/dL (the Endocrine Society's, from the harmonized data), confirmed on a second morning sample and accompanied by symptoms, is the usual basis for a diagnosis of hypogonadism. The number alone is never sufficient, and no guideline treats a single reading or a reading without symptoms as a diagnosis.

The gap between 264 and 300 is where most of the argument lives, and it is small compared with the variation introduced by assay, time of day and SHBG. A man at 280 ng/dL with normal SHBG, a normal calculated free testosterone and no symptoms is not hypogonadal under either guideline. A man at 320 with high SHBG, a low free testosterone and the EMAS triad of sexual symptoms may be. This is why the guidelines describe a process rather than a cut-off.

Be wary of clinics that move the line. A threshold set at 400 or 500 ng/dL, or a "suboptimal" range that captures half of healthy men, exists to qualify customers, not to diagnose patients. The Endocrine Society guideline that sets the standard is Bhasin et al., 2018.

No. "Low for your age" places you below the average for men of your decade; by definition half of healthy men are there, and a man in the lower part of the normal range with no symptoms is well. "Clinically low" means a level below the diagnostic threshold, confirmed twice in the morning, with symptoms that fit and no better explanation. The first is a statistic; the second is a diagnosis.

The conflation is commercially useful and medically harmful. A man told his 420 ng/dL is "low for his age" may be sold treatment he does not need, with the risks that entails, for a benefit that trials have not demonstrated in men with normal levels. The reverse confusion, assuming a level "normal for age" cannot be deficient, misses the older man whose high SHBG hides a genuinely low free testosterone.

The useful questions are therefore not comparative but clinical: are my levels genuinely below threshold on proper testing, do I have symptoms that fit, is my free testosterone low, and is there a reversible cause? Those are the questions a clinician answers and a chart cannot.

The symptoms are broadly the same at every age: reduced libido, fewer morning erections and erectile difficulty, fatigue, low or flat mood, reduced motivation, loss of muscle and strength, increased body fat, and sometimes breast tenderness, reduced body hair, hot flushes or smaller testes. What changes with age is how likely other causes are and how specific the symptoms are. The sexual symptoms are the most specific; fatigue and low mood are the least.

In EMAS, only the three sexual symptoms tracked testosterone levels consistently enough to be used in a definition; physical and psychological symptoms were associated far more weakly, because they have so many other causes. That is a useful hierarchy for a man assessing himself: a clear fall in sexual desire and function points more strongly toward testosterone than tiredness alone does.

In a younger man, the same symptoms more often reflect a specific cause worth finding. In an older man they are more often multifactorial, and sorting them requires the full evaluation rather than a single blood test. Either way, symptoms are a reason to test, not a substitute for testing. Our low testosterone symptoms guide covers each in detail.

With two fasting morning total testosterone measurements on separate days, interpreted alongside symptoms and, where indicated, SHBG and calculated free testosterone, LH and FSH to localize the problem, prolactin, a complete blood count, and a review of medications, sleep, weight and general health. The Endocrine Society and the American Urological Association both specify this sequence. A diagnosis made from one afternoon reading or a symptom questionnaire alone does not meet it.

• Two morning samples. Fasting, before 10 a.m., on different days, when the man is well.
• Symptoms. Recorded, with attention to the specific sexual symptoms.
• SHBG and free testosterone. When the total is borderline, the man is older or obese, or symptoms and number disagree.
• LH and FSH. High values point to the testes; low or normal values point to the pituitary and prompt a prolactin and sometimes imaging.
• Other causes. Medications, sleep apnoea, iron overload, chronic illness.
• Baselines for treatment. Haematocrit, PSA in men over 40, blood pressure.

Telehealth can deliver all of this properly, through partner labs or at-home morning draws and a clinician who reviews the results, and the better online clinics do. The test of a clinic is whether its process matches the list above. Our guide to how to get TRT online describes what a sound remote diagnosis looks like.

Because testosterone follows a daily rhythm that peaks in the early morning and falls through the day, by 20 to 30% in younger men, and because every reference range was built from morning samples. An afternoon value is both lower than the man's true peak and compared against a yardstick that does not apply to it. The guidelines specify samples before 10 a.m., fasting, and repeated.

The rhythm flattens with age but persists, and other short-term factors stack on top of it: a bad night's sleep, a heavy meal, a hard workout the day before, an acute illness, all lower the reading temporarily. Two morning samples on separate good days are the minimum that smooths this noise enough to trust.

Men regularly receive a single afternoon result, are told they are low, and start down the treatment path on the strength of it. Proper morning testing frequently tells a different story. If you have been tested once, at the wrong time, the first step is simply to be tested properly.

Expect it, yes: the decline is a normal part of ageing and a level that would be low for a 25-year-old is often unremarkable for a 70-year-old. Treat it automatically, no: the FDA label, the Endocrine Society and the American Urological Association all advise against prescribing testosterone for a lower level with age in the absence of symptoms and a confirmed diagnosis. Treat a confirmed, symptomatic deficiency in an older man, with the extra monitoring his age warrants, yes.

The evidence supports both halves. Trials of testosterone in older men with mildly low levels and nonspecific symptoms have shown modest benefits at best; the T Trials in men over 65, for example, found improvements in sexual function and some measures of mood and anaemia, little effect on walking distance or cognition, and an increase in coronary plaque volume in a small imaging substudy that TRAVERSE later did not translate into events. Older men with clear hypogonadism, on the other hand, gain the same symptomatic relief as younger men.

The judgement is therefore individual and should be made with a clinician who will check the prostate, blood count and blood pressure more closely than in a younger man, as is TRT safe? describes.

Yes, and when they do it usually has a specific cause: Klinefelter syndrome or other primary testicular conditions, pituitary disease including prolactinomas, prior anabolic steroid use, opioid medication, severe obesity, untreated sleep apnoea, haemochromatosis, or the after-effects of chemotherapy, testicular injury or mumps orchitis. Because age-related decline has barely started before 40, a low level in a young man is a reason to look harder, not to prescribe faster.

The investigation is more extensive than in an older man and more likely to change the plan. LH and FSH separate testicular from pituitary causes; prolactin and, if raised, pituitary imaging find tumours that are treated differently; a steroid history explains a suppressed axis that may recover; a karyotype confirms Klinefelter syndrome, which has implications beyond testosterone. Fertility is almost always relevant and must be discussed before any testosterone is prescribed, since TRT suppresses sperm production.

A young man with two low morning readings and symptoms deserves this workup and a conversation about fertility-sparing options such as enclomiphene or hCG. A young man with one borderline reading and an online form deserves a second test. Our guides to TRT and fertility and enclomiphene versus TRT cover the choices.

There is no ideal number. For a man not on treatment, good means inside the normal range for his lab and free of deficiency symptoms, wherever in that range he sits. For a man on treatment, guidelines target the mid-normal range, roughly 400 to 700 ng/dL, measured at the right point in the dosing cycle, because that is where symptoms resolve and the risks that rise with higher levels stay in check. Above the normal range there is no added benefit and there are added harms.

The belief that more is better drives most of the harm in this field. Within the normal range, a man at 500 is not less healthy than a man at 800. Above it, haematocrit, blood pressure, estradiol conversion and the suppression of natural production all climb, and the benefits the man was seeking do not. Legitimate therapy targets normal and symptom-free; chasing the top of the range, or beyond it, is performance use under another name. Our TRT versus steroids guide draws the distinction, and TRT benefits describes what replacement does and does not deliver.

To a degree, and most reliably in men whose levels are lowered by something modifiable. Losing excess fat is the strongest lever: in cohort data the rise in testosterone scales with the weight lost. Sleeping seven to eight hours and treating sleep apnoea, resistance training, limiting alcohol, managing chronic stress, correcting a genuine vitamin D or zinc deficiency, and reviewing medications that suppress testosterone all help. None of them will overcome a structural deficiency, and over-the-counter "testosterone boosters" have no good evidence behind them.

• Weight. The largest and best-supported effect, especially from visceral fat.
• Sleep. Quantity, quality and apnoea treatment.
• Training. Progressive resistance work supports body composition and modestly supports testosterone.
• Alcohol and stress. Heavy drinking and chronic cortisol elevation both suppress the axis.
• Medications. Opioids and glucocorticoids in particular, reviewed with the prescriber.

For men with a true deficiency these habits make treatment work better rather than replacing it. For men whose decline is lifestyle-driven they are the treatment. Our guide to natural ways to support testosterone covers the evidence for each.

When you have persistent symptoms that fit, especially the sexual ones, or a medical situation that calls for it: infertility, osteoporosis in a man under 70, a pituitary problem, long-term opioid or glucocorticoid use, HIV with weight loss, or a history that suggests Klinefelter syndrome. Routine testing of men without symptoms is not recommended by any major guideline, because a number in a man with no complaint rarely changes anything and often creates a worry that did not exist.

The reason screening is discouraged is practical. Finding a 310 ng/dL in an asymptomatic 52-year-old does not make him ill, but it does invite treatment he does not need and anxiety about a value that is normal for him. Testing is most useful when it answers a question the man has already asked with his symptoms.

When you do test, do it properly: morning, fasting, repeated if low, and interpreted with your symptoms. If the symptoms have an obvious other explanation, a period of poor sleep, a new medication, a bereavement, address that first. Our signs of low testosterone guide helps decide whether testing is warranted.

Read the number against the reference range printed on your own report, check that it was a morning sample taken when you were well, look at SHBG and free testosterone if they are there, and set the whole thing against your symptoms. One low reading is a prompt to repeat, not a diagnosis. A borderline reading with high SHBG may be more significant than it looks; a borderline reading with no symptoms is usually not significant at all.

• Your lab's range, not a chart from elsewhere.
• Timing and condition. Morning, fasting, not during illness.
• Free testosterone, calculated from total, SHBG and albumin if the direct value is from an immunoassay.
• Symptoms, with weight given to the specific sexual symptoms.
• Repeat before concluding anything.

The common mistakes are predictable: treating one afternoon reading as a verdict, comparing the number with an unsourced table, assuming below-average means deficient, and ignoring a confirmed low result because it is "only a little" under the line. A clinician who understands assays, SHBG and symptoms is the right interpreter of a borderline result; the complete guide to TRT describes what happens next if the result is genuinely low.

Within the normal range, no: there is no good evidence that men at the top of the range are healthier than men in the middle, and the range is wide because a wide band of levels is compatible with health. Above the normal range, the answer is clearly no: supraphysiological testosterone, from excessive therapy or anabolic steroids, raises haematocrit and clotting risk, blood pressure and cardiovascular strain, suppresses fertility, and delivers no proportional benefit.

The observational studies that find low testosterone associated with worse health are measuring, in large part, the illnesses that lower testosterone. They do not show that pushing a normal level higher improves anything, and the randomized evidence in men with normal or mildly low levels does not either. Testosterone is not a score.

The constructive frame is the one that runs through this guide: aim to be in range and free of symptoms, address the health factors that lower levels, and treat a genuine deficiency properly. Men who do that have captured the benefit; men who chase a higher number have captured only the risk.

It provides context, not the decision. Knowing the normal curve stops a man from treating a level that is simply right for his age, and stops a clinician from dismissing a genuinely low level in a young man as nothing. The decision itself rests on the diagnostic process: confirmed low levels, symptoms, free testosterone where relevant, a search for reversible causes, and an individual weighing of benefit against monitored risk.

Age shifts the emphasis inside that process. In a young man it pushes toward finding a cause and protecting fertility. In an older man it pushes toward free testosterone, toward other explanations for nonspecific symptoms, and toward closer monitoring of the prostate, blood count and blood pressure if treatment goes ahead. At every age it argues against treating a number.

The man who reads this guide and still wonders whether he should be treated has the right next step: a properly conducted test and a clinician's evaluation. Our safety guide, the guide for beginners and the clinic comparison cover what comes after.

Not for the decline alone. Every major guideline and the FDA label advise against testosterone for a lower level with age in a man without symptoms and a confirmed diagnosis, because the benefit is unproven and the risks are real. It is appropriate for an older man with confirmed, symptomatic hypogonadism after reversible causes have been addressed, with monitoring matched to his age. The difference between those two men is the whole of the question.

The case against treating the number rests on the trial record: in older men with mildly low levels, testosterone produced modest or inconsistent benefits, and the risks that need monitoring are more common with age. The case for treating the diagnosis rests on the same record read the other way: men who met the criteria improved, and TRAVERSE showed that properly selected older men could be treated without an increase in major cardiac events.

A clinic that treats every age-related dip is getting this wrong in one direction; a doctor who refuses to consider a symptomatic 68-year-old with a total of 190 ng/dL is getting it wrong in the other. The guidelines sit between them, and so should the decision.

Confirm it properly, with a second fasting morning sample. Ask honestly whether you have symptoms, and which ones. Address what is reversible: weight, sleep, alcohol, medications. If a confirmed low level and real symptoms persist, get a full evaluation that includes SHBG, free testosterone, LH, FSH and a search for a cause. Then, and only then, decide about treatment with a clinician.

• Repeat the test before you do anything else; one reading is noise.
• Take the symptom inventory seriously, weighting the sexual symptoms.
• Fix the modifiable factors and retest in three to six months if the picture allows.
• Get the full workup if it persists.
• Decide with a clinician, weighing fertility, risks and monitoring.

What to avoid at either end: ignoring a confirmed, symptomatic deficiency because you are "getting older", and buying testosterone, online or from a gym, on the strength of a single number and a marketing page. If the evaluation leads to treatment, the clinic comparison and provider reviews show how legitimate clinics handle testing, clinician review and monitoring. If it leads somewhere else, that is a good outcome too.

Testosterone peaks in the late teens and twenties, with a harmonized reference range of 264 to 916 ng/dL and a median near 531 for healthy men under 40, and falls by about 1% a year in total and 2% a year in free testosterone from the thirties onward. There is no reliable standardized chart of decade averages. "Low for your age" is not a diagnosis. A diagnosis needs two low morning readings, symptoms that fit and no better explanation, and treatment is for that diagnosis, not for ageing.

• The shape of the curve is well established; the exact number for your age is not, and your own lab's range is the one that applies.
• Much age-related decline is health-related decline, and weight, sleep and medications are where to look first.
• Free testosterone matters more as SHBG rises with age, and more when the total is borderline.
• Diagnosis is a process, morning, twice, with symptoms, not a cut-off.
• Treatment is for genuine, symptomatic deficiency, with monitoring, and never for a higher number.

This guide is general information, not medical advice. The reference figures come from Travison et al., Journal of Clinical Endocrinology and Metabolism, 2017 (harmonized range), Harman et al., JCEM 2001 (rate of decline), Wu et al., New England Journal of Medicine, 2010 (late-onset hypogonadism criteria and prevalence), and the Endocrine Society's 2018 clinical practice guideline. Reference ranges vary by laboratory and assay; interpret your result with a licensed clinician. For treatment, see the complete guide to TRT and is TRT safe?.