Testosterone

Testosterone is the most over-prescribed hormone in American medicine right now, and it’s not close. Walk into a strip-mall “low T clinic” with a pulse and a complaint about being tired and you’ll walk out with a prescription for testosterone cypionate (the long-acting injectable form) and a follow-up appointment to buy more of it. Nobody drew labs at 7 AM fasting, nobody asked about your sleep, nobody checked whether you snore loud enough to wake your wife. The business model requires that you start the medication and keep paying for it, and the business model is very good at what it does.

The clinical reality is messier than either the marketing or the establishment guidelines admit. Sometimes the answer is your sleep is wrecked, sometimes the answer is too much alcohol and a job that’s eating you alive, and sometimes the answer really is that your nuts don’t work like they used to and the family jewels aren’t shining the way they were a decade ago. Figuring out which of those you’re dealing with is the whole game.

Some guys have hypogonadism by the official numbers (the medical word for when your body isn’t producing enough testosterone, either because the testicles aren’t doing their job or because the brain signals telling them to do their job aren’t getting through) and they benefit a lot from replacement. Plenty more guys have testosterone that’s officially “in range” but actually low for their age, which is a real problem that mostly goes untreated because the establishment cutoff was set decades ago for a different population and is set way too low. And a third group of guys feel like garbage for reasons that aren’t testosterone at all, where starting injections without checking means missing the actual problem and putting them on a lifelong medication for no good reason. The field gets all three wrong in different directions on a daily basis.

What “normal” actually means, and why the official floor is way too low

When the Endocrine Society says you need a morning total testosterone below 300 ng/dL to call it low, understand where that number comes from. It’s the bottom 2.5 percent of men, statistically. “Low enough to be in the worst tail” is what gets you the diagnostic label. It does not mean “healthy for your age.” Those are different questions, and the way the label gets used in practice mostly ignores the difference.

Look at the same data broken out by age and the picture gets clearer. Healthy non-obese men aged 20 to 24 sit in a middle band of roughly 410 to 560 ng/dL, and for 25 to 29 year olds that band runs about 413 to 575. The low-T cutoff that falls out of that data for a man in his mid-20s is around 413 ng/dL, not 300. A 25-year-old at 350 is technically “in range” by the Endocrine Society number but he’s well below the normal range for his age, which is a clinical signal, not a normal finding.

It gets worse when you look at how the population baseline itself has dropped. Testosterone has been declining in American men, generation over generation, and the paper that put a number on that decline is Travison’s 2007 study, which is literally titled a population-level decline in serum testosterone levels in American men. Median testosterone fell from 501 ng/dL in the late 1980s to 391 ng/dL by the early 2000s, roughly a 22 percent drop across about fifteen years, and the authors found it held up even after they accounted for aging and weight gain, so this isn’t just the men getting older or heavier. The mechanisms are still argued over and this is an observed population trend rather than a settled story, but the direction’s hard to ignore, and I see it in the chair too: the baseline numbers a healthy young guy walks in with keep drifting down compared to what the men a generation older were running at the same age. The 300 threshold has been applied to a population whose middle of the bell curve has slid down toward it. “Low normal” means something very different now than it did to your grandfather.

My take, after seeing a lot of guys: a healthy 30-year-old should sit comfortably between the high 500s and high 900s. A healthy 40-year-old between the mid-400s and mid-800s. The 300 floor is roughly the right number for a 70-year-old. For everyone else it’s several decades of biology too low.

This doesn’t mean a 30-year-old at 350 ng/dL automatically needs replacement testosterone (TRT, the umbrella term for prescribed testosterone in any of its forms… injections, gels, pellets, patches). It means he automatically needs a workup. Those are not the same thing.

What a real low-T diagnosis looks like

The Endocrine Society guideline is not subtle about this. You need two separate morning total testosterone levels, drawn fasting before 10 AM, both below the 300 ng/dL threshold. You need actual hypogonadal symptoms, which means specific things: low libido, erectile dysfunction, loss of morning erections, loss of body hair, gynecomastia (breast tissue growth in men), decreased muscle mass, decreased bone density. Not “I’m tired,” not “I don’t feel like myself.” Tired is not a hypogonadal symptom, tired is what happens when you’re a 47-year-old man in 2026.

The labs have to be morning-drawn and fasting because testosterone has a daily cycle. A guy who’s 450 ng/dL at 8 AM is going to be 280 ng/dL at 4 PM, and the 4 PM number means nothing. Afternoon draws are one of the most common ways men get falsely diagnosed with low T… the clinic gets a low number, calls it pathology, starts the injections, and the number was never pathological in the first place.

The workup isn’t gatekeeping. It’s the difference between treating something we can fix and slapping a band-aid on a bullet hole.

Once you do have two confirmed low morning levels, the next step is figuring out why. You check LH and FSH (the brain hormones that signal the testes to make testosterone and sperm). High LH with low T means primary hypogonadism (the testes themselves aren’t producing). Low or normal LH with low T means secondary hypogonadism (the brain isn’t signaling them to). You check prolactin (another pituitary hormone, which can suppress testosterone when it’s elevated), you check iron studies for hemochromatosis (an iron-overload condition that can wreck the pituitary), you image the pituitary if there’s a reason. None of this happens at a low-T clinic, which doesn’t care which kind you have because they sell the same thing to either.

Those are the strict textbook criteria, and a guy who meets them clearly has hypogonadism. The harder cases, and honestly more of my cases, are guys whose numbers are “in range” by the Endocrine Society cutoff but clearly low for their age. They’re symptomatic, they don’t fit the strict checklist, and they’re real candidates for treatment if anyone is willing to look at them as more than a single number against a single threshold. The strict criteria catch the textbook cases and they miss a lot of guys with real, age-inappropriate low T who would benefit from treatment if they had a prescriber who didn’t just stop reading at “above 300.”

The four things that mimic low T

Here’s what actually causes the symptoms most men show up with, in rough order of how often it turns out to be the real culprit:

Sleep apnea. A 50-year-old man with a 17-inch neck and a snoring problem will have low energy, low libido, brain fog, weight gain, and yes, suppressed testosterone, because apnea suppresses the whole brain-to-testes signaling axis. Treat the apnea and the T often comes up on its own. Start him on TRT instead and you’ll make his apnea worse, because exogenous testosterone independently worsens sleep apnea. This is the single biggest miss in the whole industry.

Depression. Major depression suppresses libido, motivation, energy, and yes, often testosterone. The symptom overlap with low T is nearly total. A guy comes in saying he has no drive, no interest in sex, feels like he’s underwater, and the question is whether that’s a hormonal issue or whether he’s been clinically depressed for three years and never told anyone. SSRIs help the second guy. Cypionate does nothing for him.

Alcohol. Two or three drinks a night, every night, for years. Alcohol directly suppresses testosterone production and ruins sleep quality on top of it. Quitting drinking can pull testosterone up meaningfully with zero other intervention. Nobody at the TRT clinic asked how much you were drinking.

Deconditioning and weight gain. Body fat converts testosterone into estradiol (the main female sex hormone, which men make a small amount of), so a 40-pound gain over the last five years will pull total T down meaningfully. Testosterone won’t fix that. Losing the 40 pounds will.

The reason for checking these four isn’t to deny you TRT, it’s because if any of them are present and we don’t address them, the TRT either won’t work well or will actively make some of them worse (apnea especially). Sometimes fixing the underlying issue brings your testosterone up enough on its own that you don’t need lifelong injections, which is the best outcome because lifelong injections aren’t a thing anyone should start lightly. And sometimes you’ve cleaned up the apnea and the alcohol and the depression and your testosterone is still low for your age, in which case the TRT we start has a much better shot at actually working than if we’d just jumped to it on day one.

If you haven’t ruled out sleep apnea, alcohol, depression, and weight before starting TRT, you haven’t ruled out anything.

What the workup catches that the clinic misses

This is where the workup pays off. Picture a guy who’s been getting told for a year or two by his primary care that his fatigue is just stress. He shows up convinced he’s depressed. He isn’t, he’s actually hypogonadal, and the labs come back with morning totals in the 180 to 200 range, two draws a month apart. LH is inappropriately low-normal, which points at secondary hypogonadism… the brain isn’t signaling the testes to work. Pituitary MRI shows a small prolactinoma (a benign tumor in the pituitary gland that makes too much prolactin, which is what’s been suppressing the testosterone). Treat the prolactinoma with cabergoline (a pill that shrinks it), and the testosterone comes back up to the high 400s on its own within four months, with the fatigue and libido issues resolving alongside it. He never needed prescription testosterone. The clinic that would’ve started him on cypionate would’ve masked the prolactinoma for years, until it got big enough to start pressing on the nerves behind his eyes.

That’s the point of doing the workup. Not gatekeeping. Catching the brain tumor before it gets big enough to compress the optic nerves and tank your vision.

If you actually need it, here’s what it does and what it costs

Say you’ve been worked up properly, your numbers are below normal for your age (not just below the Endocrine Society’s one-size-fits-no-one cutoff), and you’ve talked through the trade-offs with somebody honest. The decision to go on TRT is yours. I’m a provider, not a parent. If you’ve heard the risks (the fertility issue, the cardiovascular monitoring, the lifelong commitment) and you still want the script, you get the script. I’m opinionated about how this should be done, not a gatekeeper standing in the way.

The good news first, because the rest of this section is about side effects and nobody leads with the upside: TRT works for the patient who actually has hypogonadism, and it works well. Energy, mood, libido, body composition, bone density, all of it comes back up over the first few months. A lot of guys describe it as feeling like themselves again for the first time in years. That’s what the marketing material is selling, and for the patient who actually has the condition the marketing material is selling to, the marketing material happens to be true. The catch is that “for the patient who actually has the condition” was doing a lot of work in that sentence. Worth knowing the honest version of the evidence too, because the field is genuinely split on whether TRT does anything for sexual function. The big 2024 Cochrane review pooled 43 trials and found that across the broad population studied, testosterone made little to no average difference in erectile function or sexual quality of life versus placebo, which is a useful gut-check against anybody promising it’s a magic libido switch. But then you’ve got the Testosterone Trials, a set of randomized trials in older men with genuinely low testosterone confirmed on two morning draws, and those found TRT actually did improve sexual activity, desire, and erectile function versus placebo in exactly that deficient population. So the jury’s out depending on who you pool: throw everybody into one bucket and the benefit washes out, narrow it to men who are really low and the benefit shows up. Here’s where I land. It works for the guy who’s actually deficient, and it’s oversold for the average tired guy who just wants a number bumped a little, and the reason the broad Cochrane pooling looks so flat is that the second guy is diluting the first guy’s result. The men who do best are the ones who genuinely started out deficient, not the ones chasing an above-average number.

The side effects are real, and the bro-science crowd mostly downplays them.

The fertility piece is the one younger men keep getting blindsided by. A 32-year-old who hasn’t had kids yet starts TRT at a clinic that never had the conversation, his sperm count drops to almost nothing within a year, and now he’s sitting in a fertility workup wondering what happened. If you’re under 40 and you might want kids, your options are hCG monotherapy (a hormone that keeps the testes making sperm even while you’re on TRT), clomiphene (a pill that boosts your own testosterone production rather than replacing it from outside), or banking sperm before you start. The clinic that didn’t bring this up failed you.

The cardiovascular fear has cooled off too. The big randomized safety trial (TRAVERSE, more than 5,000 hypogonadal men who already had heart disease or were at high risk for it) found testosterone gel was no worse than placebo for major cardiac events, though the testosterone group did see a few more arrhythmias, including atrial fibrillation. So it isn’t a free pass, it’s “not the heart-attack machine the old scare headlines made it out to be.”

The prostate question is less alarming than it used to be too. The old fear was that testosterone would feed prostate cancer. The data over the last fifteen years hasn’t really borne that out for men without active cancer, but you still check a baseline PSA (the prostate cancer screening blood test) in men over 40, you still do a digital rectal exam if it’s indicated, and you recheck PSA periodically. If a man already has prostate cancer, that’s a separate conversation with urology, not something a low-T clinic should be deciding on its own.

Monitoring schedule, the boring version: baseline labs (total T, free T, LH, FSH, prolactin, complete blood count, PSA if over 40, lipid panel, metabolic panel, hematocrit). At 3 months, recheck T and the red blood cell percentage. At 6 months, full panel again. Then yearly forever. Sleep study if there’s any snoring history or if hematocrit starts trending up. A prescriber who isn’t doing all of that isn’t managing the treatment, they’re just refilling a script.

The “I just want to feel better” patient

The most common version of this story is the guy in his mid-30s to late 40s who comes in saying he doesn’t feel like himself anymore. He’s tired in the afternoons, his marriage is in a rough patch, he’s gained 25 pounds, he’s not sleeping well, and a friend at the gym told him to get his T checked. He’s read enough internet to be hopeful low T is the answer.

Sometimes it is. More often he has poor sleep, too much alcohol, a depression he hasn’t named, a marriage problem he’s been avoiding, and a body that hasn’t moved meaningfully in a decade. Testosterone, even if his number is borderline, doesn’t fix any of that. It gives him about six weeks of feeling sharper, which is mostly the relief of having done something, and then he’s back to baseline, still avoiding the actual problems, except now he’s on a medication he might be on for life and growing breast tissue.

Which doesn’t mean he never gets TRT. It means we check the other things first because if we DO treat the sleep and the drinking and the depression and the deconditioning, his testosterone might come up enough on its own that he doesn’t need lifelong injections, or it might not, and either way we’ll know what we’re actually dealing with. If we work all that up and his numbers are still low for his age, then yes, we have the TRT conversation, and his decision is the deciding one.

The workup isn’t gatekeeping. It’s the difference between treating something we can fix and slapping a band-aid on a bullet hole. A clinic that hands out cypionate without one is selling a feeling for the price of a treatment.

For the broader workup logic on the sexual-function side of things (psychogenic versus blood-flow problems, medication-induced dysfunction, when to involve which specialist), see sexual health.

Sources

1. Zhu A, Andino J, Daignault-Newton S, et al. What Is a Normal Testosterone Level for Young Men? Rethinking the 300 ng/dL Cutoff for Testosterone Deficiency in Men 20-44 Years Old, J Urol, 2022;208(6):1295-1302. PMID 36282060.
2. Travison TG, Vesper HW, Orwoll E, et al. Harmonized Reference Ranges for Circulating Testosterone Levels in Men of Four Cohort Studies in the United States and Europe, J Clin Endocrinol Metab, 2017;102(4):1161-1173. PMID 28324103.
3. Travison TG, Araujo AB, O’Donnell AB, Kupelian V, McKinlay JB. A population-level decline in serum testosterone levels in American men, J Clin Endocrinol Metab, 2007;92(1):196-202. PMID 17062768.
4. Bhasin S, Brito JP, Cunningham GR, et al. Testosterone Therapy in Men With Hypogonadism: An Endocrine Society Clinical Practice Guideline, J Clin Endocrinol Metab, 2018;103(5):1715-1744. PMID 29562364.
5. Lincoff AM, Bhasin S, Flevaris P, et al. Cardiovascular Safety of Testosterone-Replacement Therapy (TRAVERSE), N Engl J Med, 2023;389(2):107-117. PMID 37326322.
6. Lee H, Hwang EC, Oh CK, et al. Testosterone Replacement in Men With Sexual Dysfunction, Cochrane Database Syst Rev, 2024;1:CD013071. PMID 38224135.
7. Snyder PJ, Bhasin S, Cunningham GR, et al. Effects of Testosterone Treatment in Older Men, N Engl J Med, 2016;374(7):611-624. PMID 26886521.