Why Testosterone Is Falling Across Generations

A man of 35 tests at 14 nmol/L and is told it is normal. It is. His father at 35 would have tested closer to 19, and his grandfather higher still.

None of them would have called it a problem, because each was measured against the men of his own time. That is how reference ranges work: they describe what is typical, not what is healthy. And what is typical has been moving.

For two decades this was a claim made mostly by people selling something. In 2025 it stopped being that, because someone finally counted properly.

What the largest analysis found

A systematic review published in 2025 pulled together every study measuring testosterone in healthy men from 1971 to 2024. It ended up with 1,256 papers, 1,504 separate study groups, and 1,064,891 men, average age 42.

The finding was a statistically significant decline in testosterone across the decades, and it held after adjusting for the size of each study, the men's ages, their body mass index, and the assay used to measure them.

That last adjustment matters more than it sounds. The obvious objection to any historical trend is that the measuring instrument changed, and older assays were less precise. Adjusting for assay type removes that explanation. The decline survives it.

Two secondary findings are what make this genuinely interesting.

The two numbers that rule out the easy explanations

Body mass index showed no trend. Within this population of healthy men, BMI did not rise across the study period, yet testosterone fell anyway.

This is awkward for the standard explanation. Obesity lowers testosterone, obesity has risen, therefore testosterone has fallen. That reasoning is sound and it explains part of the picture, but it cannot explain a decline in men whose BMI did not change.

Luteinising hormone fell too. This is the finding that should get the attention, and almost nobody reports it.

LH is the signal your pituitary sends to your testes telling them to produce testosterone. When the testes fail, the brain notices and shouts louder: testosterone drops and LH climbs. That is primary hypogonadism, and it is what testicular damage looks like on a blood test.

Here, both fell together. The brain is not shouting louder. It is signalling less.

That pattern points upstream, to the hypothalamus and pituitary rather than the testes. Whatever is driving this is affecting the instruction, not the factory. What causes low testosterone explains why that distinction decides almost everything about a man's individual case, and it applies at population scale too.

How much, and over what period

The 2025 review establishes direction rather than a single headline figure, because it pools studies with different populations and methods.

Earlier work gave numbers. The Massachusetts Male Aging Study, which followed the same men over time from the late 1980s, found a decline of roughly one per cent per year that was independent of ageing. A man of 60 in 2004 had substantially lower testosterone than a man of 60 in 1988.

The pattern has since been documented in young men too. Analysis of American men aged 15 to 39 found mean total testosterone falling across successive cohorts, in a group where age-related decline should not apply at all.

So this is not older men declining faster. It is the whole distribution shifting down.

Why the ranges hide it

Here is the practical consequence, and it is the reason this matters to you rather than to epidemiologists.

A laboratory reference range is built by measuring a sample of men and taking the middle 95 per cent. It describes the population it was drawn from. If that population's testosterone falls, the range falls with it, and every man inside it continues to be told he is normal.

Normal is a moving target, and it has been moving in one direction.

This is why the phrase "your testosterone is normal" carries less information than it sounds like it does, and why a normal blood test can sit alongside genuine symptoms. It is also why testosterone levels by age is worth reading alongside this: that article covers what happens to one man over his life, while this one covers what has happened between generations.

What is actually driving it

Honest answer first: nobody knows, and any article that tells you otherwise is selling a solution. Several candidates have real evidence behind them, and they are not mutually exclusive.

Body composition, but not the way you would think

Population BMI has risen substantially since the 1970s, and obesity lowers testosterone through aromatase in fat tissue. This explains a share of the decline.

But BMI does not distinguish muscle from fat, and men have on average become less muscular and fatter at the same weight. A man of 80 kilograms in 1975 and a man of 80 kilograms today are not carrying the same tissue. Why your belly fat will not budge covers the loop this creates.

Sleep

Most of your daily testosterone is produced during deep sleep. Average sleep duration has fallen across the same decades, and sleep quality has arguably fallen further with artificial light and screens.

Restrict healthy young men to five hours a night for one week and daytime testosterone drops ten to fifteen per cent. That is one week. Consider what several decades of a shorter, more fragmented average does at population scale.

Obstructive sleep apnoea has also become more common, and it is the most missed cause in individual men. Sleep apnoea and testosterone covers what treating it does and does not do.

Physical activity

Occupational activity has collapsed. Jobs that involved walking, lifting and standing have been replaced by jobs that involve sitting, and a gym session does not fully replace eight hours of movement.

Resistance training in particular supports testosterone through body composition, and it is not what most men's days now contain. Does lifting increase testosterone covers what it does and does not do acutely.

Endocrine-disrupting chemicals

Exposure to compounds that interact with hormone receptors has risen enormously since the 1970s. The mechanism is plausible, regulatory concern is legitimate, and the timing fits.

What is missing is proof that this specific exposure causes this specific population decline. It remains a strong hypothesis rather than an established cause, and it is worth saying plainly, because it is the explanation most heavily promoted by people selling detoxes. How to lower oestrogen in men covers where sensible precaution ends and the panic begins.

Medication

Prescribing of drug classes that suppress testosterone has risen substantially: opioids, some antidepressants, and others. Opioid prescribing in particular grew dramatically over the study period, and sustained opioid use suppresses the pituitary signal profoundly.

This one deserves more attention than it gets, partly because it fits the LH finding: these drugs act upstream, exactly where the data points.

What this does not mean

It does not mean you personally have low testosterone. Population averages say nothing about you. Plenty of men today test higher than their fathers did.

It does not justify treatment on the basis of the trend. "Everyone's is falling" is not a diagnosis. Testosterone therapy requires confirmed, symptomatic deficiency, and the case for it rests on your numbers rather than on a graph. TRT before and after covers what treatment actually changes.

It does not mean the decline is irreversible in an individual. The population trend is driven largely by factors that are modifiable in one person, which is the genuinely encouraging part of this.

And it does not mean supplements address it. If sleep, activity, body composition and medication are the drivers, nothing in a capsule touches them. Supplements that waste your money covers what is being sold on the back of this trend.

What to do with it

The useful response is not alarm. It is a different attitude towards the word normal.

Get the number, and get it properly. Two morning samples on separate days, with SHBG, free testosterone, LH and FSH. LH is the one most often omitted and, given everything above, the one that tells you most. The blood test guide covers what to ask for.

Judge it against how you feel, not only against the range. A man at the bottom of a range that has itself been sliding is in a different position from a man at the bottom of a stable one.

Address the drivers you control. Sleep, visceral fat, resistance training, alcohol and medication review. These are the same factors implicated at population scale, and they are the ones that respond in an individual. The 30-day reset is the structured version.

Retest after a real attempt. Three to six months, not three weeks.

The honest summary

Testosterone has fallen across generations, the largest analysis available confirms it, and the decline survives adjustment for age, weight and measurement method.

The most informative detail is that LH fell alongside it. This is not testes failing across a population. It is a quieter signal from above, which points at sleep, activity, body composition and medication rather than at anything happening below the waist.

What it should change is not what you take, but how much weight you give the word normal on a lab report. That range describes the men around you. It was never a statement about what your body is capable of.

Common questions

Is testosterone really declining in men?

Yes. A 2025 systematic review covering 1,064,891 men across studies from 1971 to 2024 found a significant decline over time, holding after adjustment for age, BMI and the assay used.

How much has it fallen?

Estimates vary by population and method. The Massachusetts Male Aging Study found roughly one per cent per year independent of ageing, meaning a man today has substantially lower levels than a man of the same age decades ago.

Is it just because men are heavier now?

Only partly. Obesity lowers testosterone and does explain some of it, but the 2025 review found no BMI trend within its population while testosterone still fell. Weight is a contributor, not the whole answer.

What does the LH finding mean?

LH is the pituitary's signal to the testes. If testes were failing, LH would rise. Instead it fell alongside testosterone, pointing to something affecting the signal from the brain rather than production in the testes.

Does this mean my testosterone is low?

No. Population trends say nothing about an individual. It does mean a "normal" result is less reassuring than it sounds, because the range itself has been drifting downwards.

Can I do anything about it?

For yourself, yes. The suspected drivers, sleep, physical activity, body composition and certain medications, are largely modifiable in an individual even though they are hard to shift at population level.

Sources: Santi D, Spaggiari G, Furini C et al., Temporal trends in serum testosterone and luteinizing hormone levels indicate an ongoing resetting of hypothalamic-pituitary-gonadal function in healthy men: a systematic review, J Endocrinol Invest 2025 · 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 · Lokeshwar SD et al., Decline in serum testosterone levels among adolescent and young adult men in the USA, Eur Urol Focus 2021;7(4):886-9 · Leproult R, Van Cauter E, Effect of 1 week of sleep restriction on testosterone levels in young healthy men, JAMA 2011;305(21):2173-4 · Daniell HW, Hypogonadism in men consuming sustained-action oral opioids, J Pain 2002;3(5):377-84 · Educational only, not medical advice.

Keep reading: Testosterone levels by age · What causes low testosterone · Normal blood test, still low testosterone · Sleep apnoea and testosterone · Blood test: what to check · Free 2-minute testosterone test