Low Testosterone: What the Evidence Supports and What It Does Not

Key points#

• A diagnosis of low testosterone requires two things together: consistent symptoms and a low morning fasting total testosterone confirmed on a repeat test. A single number, or symptoms alone, is not a diagnosis.
• A modest, gradual decline in testosterone with age is normal. Marketing that treats every tired, low-libido man as a candidate for treatment runs well ahead of the evidence.
• The clearest, most reproducible benefit of treatment in men with confirmed deficiency is in sexual function, with a smaller effect on mood. Gains in energy, strength, and vitality are limited or unproven.
• The large randomized TRAVERSE trial found testosterone gel noninferior to placebo for major cardiac events in men at cardiovascular risk, but it flagged higher rates of atrial fibrillation, pulmonary embolism, and acute kidney injury. Reassuring, not blanket.
• Testosterone is not a fertility aid or a wellness supplement. It can suppress sperm production, and it is not FDA-approved for low testosterone due to aging alone.
• Honest care means treating confirmed deficiency, monitoring on a schedule, and being candid about what treatment can and cannot change.

What testosterone actually does, and how levels change with age#

Testosterone is the main male sex hormone, and its everyday jobs are less dramatic than the advertising suggests. It supports libido and erectile function, helps maintain bone density and muscle mass, drives the production of red blood cells, and plays a part in mood and a general sense of well-being. When the level is genuinely and persistently low, several of these can slip at once, which is what makes real deficiency worth treating.

Levels also fall with age, and this is where clear thinking matters most. In most men, testosterone declines gradually and modestly across adulthood. That downward drift is a normal feature of aging, not a diagnosis. A level lower than it was at twenty-five is expected at sixty, and by itself it is not a disease that needs correcting.

Low testosterone comes in two broad flavors. Primary hypogonadism means the problem is in the testes themselves. Secondary hypogonadism means the signal from the pituitary gland or hypothalamus in the brain is weak, so the testes are never properly told to work. The distinction changes what a clinician looks for next, and the Endocrine Society guideline treats finding the cause as part of the diagnosis, not an afterthought.

The symptoms are real, but they are nonspecific#

The symptoms attributed to low testosterone are common and genuinely bothersome: low libido, difficulty with erections, fatigue, low mood, loss of muscle, and trouble concentrating. The catch is that almost none of them point cleanly at a hormone. Notice how much of that list overlaps with poor sleep, depression, thyroid disease, the side effects of common medications, obesity, and ordinary aging.

Fatigue is the clearest example. It is one of the least specific symptoms in all of medicine, and a man who sleeps five hours a night, carries excess weight, and feels flat is describing several problems, of which a hormone level is only one.

Some symptoms do carry more diagnostic weight than others. Reduced morning erections, clearly diminished sexual desire, and loss of body hair are more specific to low testosterone than tiredness or low mood alone. That is why a careful clinician weighs the pattern, not just the presence of a complaint. This is the exact gap that aggressive marketing exploits: it takes a nonspecific midlife slump and maps it onto a single hormone that happens to have a convenient treatment.

When testing makes sense, and when it does not#

Testing is not screening. Guidelines do not recommend measuring testosterone in men who feel well, and a routine check in an asymptomatic man tends to create problems rather than solve them. The right trigger is symptoms and signs that actually fit deficiency.

When a test is warranted, how it is done matters as much as whether it is done. Testosterone follows a daily rhythm and is highest in the morning, so the sample should be drawn in the morning, ideally fasting, using a reliable assay. Just as important, a single low result does not settle anything. Levels fluctuate, assays vary, and a passing illness can push a number down, so a low reading should be confirmed on a repeat measurement before anyone says deficiency.

There is a further wrinkle. Most testosterone in the blood is bound to a carrier protein called sex hormone binding globulin (SHBG), and only the unbound, or free, fraction is active. When SHBG is altered (by obesity, thyroid disease, or age) or when the total sits right at the border, measuring free testosterone by equilibrium dialysis or a validated calculation gives a truer picture. And once a genuinely low level is confirmed, a low result should prompt a look for the underlying cause rather than a reflex prescription. The Endocrine Society guideline lays out this stepwise approach precisely to keep false positives from turning into unnecessary treatment.

What the evidence supports treatment can do#

Here is the honest version of the upside, drawn from the best randomized data we have. The coordinated Testosterone Trials enrolled men aged 65 and older with confirmed low levels and measured several outcomes at once. Sexual function improved by a moderate amount. Mood improved by a small amount. Vitality and walking distance did not meaningfully improve.

That result is worth sitting with, because it is more specific than any advertisement. The strongest and most reproducible benefit of testosterone therapy, in men who genuinely have low levels, is in sexual function. The effect on mood is real but smaller. The effects on energy, thinking, and physical performance are modest at best and, in these trials, largely absent.

None of this makes treatment pointless. For a man whose confirmed deficiency is dragging down his sexual health, a moderate, evidence-backed improvement in exactly that domain is a good reason to treat. What the evidence does not support is the promise of a total transformation. Setting that expectation straight before starting is part of doing this well.

What the evidence does not support, and the honest limits#

Testosterone is not established as a general anti-aging treatment. The FDA's current testosterone information, updated in June 2026, states that approved products are for men who have low testosterone together with an associated medical condition; no approved product is indicated for a low level without such a condition. That is why a number, symptoms, and the search for an underlying cause all belong together before treatment is considered.

It does not reliably restore youthful energy or strength. It is not a wellness product, and it is not a general fix for feeling run down. There is also a specific, often-overlooked harm: testosterone therapy can suppress the body's own production of testosterone and sperm, so it is not appropriate for a man who wants to father a child in the near term. For couples trying to conceive, starting testosterone can be exactly the wrong move.

Two more limits are worth naming. Benefits stop when treatment stops, so this is a long-term commitment rather than a short course. And the over-the-counter testosterone boosters sold as a shortcut generally lack good evidence that they raise testosterone or help symptoms, while facing far less oversight than prescription products. If symptoms are real, an evaluation beats an unregulated bottle.

Risks, safety signals, and monitoring#

The safety picture improved with a large trial designed specifically to test it. TRAVERSE randomized men who had, or were at high risk of, cardiovascular disease to testosterone gel or placebo. For major adverse cardiac events (a composite of cardiovascular death, heart attack, and stroke), testosterone was noninferior to placebo. A companion prostate safety analysis found no significant increase in high-grade or overall prostate cancer in this carefully screened population. Both findings are genuinely reassuring, and they push back on some older fears.

Reassuring is not the same as harmless. In the same trial, the testosterone group had higher rates of atrial fibrillation, pulmonary embolism, and acute kidney injury. Testosterone can also raise hematocrit, meaning the blood becomes thicker, which is worth watching in its own right. A responsible reading holds both halves at once: the headline cardiac fear did not materialize, and there are real, specific risks that require attention.

That is what monitoring is for. Guideline-based follow-up means rechecking symptoms to see whether treatment is helping, rechecking the testosterone level, checking hematocrit to catch blood thickening, and attending to prostate risk over time. There are also situations where testosterone should not be started: a recent heart attack or stroke, certain cancers, untreated severe obstructive sleep apnea, and a desire for near-term fertility all argue against it. The Endocrine Society guideline sets out this schedule so that treatment stays a decision you keep re-examining rather than one you make once and forget.

Talking with a clinician: questions worth asking#

Bring these to the visit. They are built to surface the real tradeoffs, not to push you toward a yes or a no.

• Were my symptoms and my level both consistent, and was the level a morning fasting sample that was repeated before anyone called it low?
• What specifically do we expect treatment to improve for me, and by roughly how much? Sexual function is the likeliest win; energy and strength are not promises.
• What is the monitoring plan, and how long before we honestly judge whether it worked?
• Could something else explain how I feel, such as sleep, mood, a medication, or weight, and have we looked?
• Do I want to preserve fertility, now or later? If so, that changes the plan.

A generalist primary care clinician is the right partner here. This is a decision that lives at the intersection of symptoms, labs, other conditions, and personal priorities, which is precisely the territory of family medicine and internal medicine. The evidence deserves careful reading rather than taking claims at face value.

Start with a plain question at your next visit: are my symptoms and my level actually consistent, and has that level been confirmed? A yes to both opens a real conversation about treatment with clear expectations and a monitoring plan. A no means the more useful work is finding out what is actually going on, which is a better outcome than a prescription that was never going to deliver what was promised.