Testosterone Is Falling in Men Under 40: What the Data Actually Shows
Population testosterone measurements really have drifted downward. What that means for any individual man is a much messier question than the telehealth ads suggest.
A 34 year old man sees an ad while scrolling at midnight. Tired? Foggy? Losing your edge? Twelve minutes and a credit card later he has ordered an at-home testosterone test. Three weeks after that he is on a subscription protocol from a telehealth clinic he has never visited, prescribed by a doctor he has never met, based on a single blood draw taken at 4 p.m. Talk of a testosterone decline in men has built an industry, and the industry is far more confident than the evidence is.
That sequence happens thousands of times a week. Underneath it sits a real observation that deserves better handling than the ad copy gives it: measurements of male testosterone in developed countries do appear to have drifted downward over the last several decades, and the drop shows up even after adjusting for age. That finding is real. Almost everything built on top of it is shakier than it looks.
None of this is medical advice, and nothing here is about doses or protocols. If you think something is wrong with your health, that is a conversation for a physician who can examine you and see your full history.
What the data on testosterone decline in men actually shows
Several long-running cohort studies in North America and Europe have compared men of the same age measured decades apart. Broadly, the more recent cohorts test lower. The commonly cited figures suggest something in the range of a low single-digit percentage decline per year across the population, which compounds into a meaningful gap over a generation.
Two things about that finding are important and usually skipped.
First, it is a population average, not a description of any individual. A shift in the mean can be produced by a large change in a subset of men rather than a small change in everyone. The distribution matters, and it is rarely reported in the headline.
Second, the decline is age-adjusted but not necessarily health-adjusted. That distinction is the entire argument, and it is where most of the interesting disagreement lives.
Why the measurement itself is messier than the number suggests
Testosterone is a genuinely difficult thing to measure well, and a lot of apparent trends partly reflect changes in how it is measured.
It swings across the day. Levels typically peak in the early morning and fall through the afternoon. The same man can produce results that differ substantially depending on whether he was drawn at 7 a.m. or 4 p.m. Guidelines generally call for morning draws for exactly this reason, and direct-to-consumer testing frequently ignores it.
It swings day to day. A single value is a snapshot of a system that fluctuates with sleep, illness, stress, and recent alcohol. Reputable clinical practice usually requires two separate morning measurements before diagnosing anything, and one low reading on its own means very little.
Assays are not interchangeable. Different laboratories use different methods, and older immunoassays and newer mass spectrometry methods do not always agree. Comparing a 1990s cohort to a 2020s cohort partly means comparing two different measuring instruments.
Total testosterone is the wrong number for many men. Most circulating testosterone is bound to sex hormone binding globulin and albumin and is not biologically available. SHBG levels shift with age, insulin resistance, thyroid function, and liver health. Two men with identical total testosterone can have quite different free testosterone. Any workup that measures only total is answering half a question.
The confounders nobody enjoys discussing
Here is the uncomfortable part of the conversation. The same decades that saw testosterone measurements fall also saw large changes in the things that most strongly influence testosterone.
Body composition. This is the biggest single driver, and it is not close. Adipose tissue converts testosterone to estradiol, and excess body fat is strongly and consistently associated with lower testosterone in men. Obesity rates in most Western countries rose sharply over the same period the hormone data declined. Studies that statistically adjust for body mass generally find the population trend shrinks considerably, though several report that some decline remains.
Sleep. Testosterone production is heavily tied to sleep, particularly the deeper stages. Short sleep studies in healthy young men have shown meaningful drops after just a week of restriction. Average sleep duration has not moved in a good direction.
Activity and muscle mass. Resistance training and general physical activity support healthy hormone function. Sedentary work has expanded across nearly every sector.
Medications and substances. Opioids suppress testosterone strongly and prescriptions rose enormously during this window. Heavy alcohol use, some antidepressants, and certain other drugs also affect levels. Prescribing patterns changed a lot over the same decades.
Metabolic health generally. Insulin resistance, poor lipid profiles, and low testosterone travel together so consistently that many endocrinologists treat low testosterone in a younger man as a signal to investigate metabolic health rather than as a standalone diagnosis.
Environmental chemical exposure gets discussed constantly in this context, usually with more confidence than the evidence supports. There is legitimate scientific concern about endocrine-disrupting compounds, and it is an active research area. There is not, at this point, a clean demonstration that a specific everyday exposure is driving population testosterone down by a specific amount. Anyone telling you otherwise with precision is going beyond what is known.
The reference range problem
When your result comes back flagged as low, you are being compared to a reference range. Those ranges are typically built from the distribution of results in a sampled population, not from a threshold at which people become symptomatic.
Follow that logic and something odd emerges. If the population being sampled is getting heavier, sleeping less, and moving less, the reference range built from that population drifts down with it. What counts as normal is partly a description of how everyone else is doing.
This cuts both ways in practice. Some men who feel genuinely bad sit just inside the range and get told nothing is wrong. Others land slightly below it, feel fine, and are told they need lifelong treatment. Ranges also vary between labs, so the same blood can be normal at one and flagged at another. A number without symptoms is not a diagnosis, and symptoms without a number are not one either.
Why the symptoms make this so easy to sell
Fatigue, low mood, poor concentration, reduced libido, and difficulty building muscle are the marketing pillars of every testosterone clinic. They are also symptoms of depression, sleep apnea, thyroid dysfunction, anemia, iron deficiency, burnout, and simply being a tired adult with a job and small children.
That overlap is the business model. A test with a low-ish result offers a clean explanation for a diffuse feeling, and a clean explanation is emotionally satisfying in a way that go to bed earlier is not. Sleep apnea in particular is worth flagging, because it is common, underdiagnosed, lowers testosterone, and produces almost exactly the symptom list the ads describe.
The trade-offs of treatment also get soft-pedaled. Exogenous testosterone suppresses the body signal to produce its own, which is the mechanism behind its effect on sperm production and fertility. It requires monitoring for red blood cell count and other markers. For many men it becomes indefinite. Those are manageable realities for someone with a genuine clinical deficiency and a real evaluation behind the diagnosis. They are a poor bargain for someone whose actual problem was six hours of broken sleep a night.
Getting a real answer about your own body
If you want to know what is actually going on, the sequence matters more than the speed.
- Get drawn properly. Early morning, ideally before 10 a.m., and not while acutely ill or after a night of drinking. Insist on it, and repeat the test on a separate day before drawing any conclusion.
- Ask for the full picture. Total testosterone alone is not enough. SHBG, free testosterone, LH, and prolactin help distinguish a testicular problem from a pituitary one, and that distinction changes what should happen next.
- Rule out the common things first. Thyroid panel, complete blood count, ferritin, and metabolic markers. If you snore, are tired despite sleeping, or your partner has noticed you stop breathing, get evaluated for sleep apnea before anything else.
- Fix the inputs for a real interval. Consistent sleep, resistance training, and losing excess fat can move levels for many men, and three to six months of honest effort is a fair test.
- See someone who is not selling the treatment. A physician who has no financial stake in the outcome, ideally an endocrinologist or urologist, will interpret a borderline number differently than a subscription clinic will.
The population trend is worth taking seriously as a public health signal. It says something about how a generation of men is sleeping, eating, moving, and being medicated, and those are problems worth naming. What it does not do is tell you anything about the specific man reading this. That still requires a proper draw, a second draw, a full panel, and a doctor who is willing to say the less lucrative thing.
Related reading
More on the basics that move health markers:
