Why Testosterone Declines With Age: What Is Actually Failing

Testosterone production runs on a three-part loop: the hypothalamus signals the pituitary, the pituitary signals the testes, and the testes produce the hormone, which feeds back to quiet the signal. Ageing degrades all three parts at once — and it separately raises the protein that binds testosterone in circulation, so a smaller share of what remains is usable.

The Baltimore Longitudinal Study of Aging tracked this in 890 men and found independent, age-invariant longitudinal declines in both testosterone and the free testosterone index.[1] Understanding which part is failing is what turns a lab number into a decision.

The loop, briefly

The hypothalamic-pituitary-gonadal axis works like a thermostat.

1. The hypothalamus releases gonadotropin-releasing hormone in pulses.
2. The pituitary answers with luteinizing hormone (LH) and follicle-stimulating hormone.
3. LH reaches the Leydig cells of the testes, which synthesise testosterone.
4. Circulating testosterone feeds back to the hypothalamus and pituitary, damping the signal.

A thermostat can fail at the sensor, at the controller, or at the furnace. Testosterone can fall for the equivalent of each — and the distinction has a name in clinical practice. The loop itself, including why external testosterone shuts it down, is set out in the HPG axis.

Failure one: the testes respond less (primary)

Leydig cell number and responsiveness decline with age. The furnace produces less heat for the same instruction.

The signature in labs is low testosterone with high or high-normal LH: the pituitary is shouting, and the testes are not answering. That pattern is primary hypogonadism, and it is the form most clearly attributable to ageing of the gonad itself.

Failure two: the signal weakens (secondary)

The hypothalamic-pituitary end also changes with age — pulse amplitude and regulation shift, so the instruction sent is weaker or less well organised.

The signature is low testosterone with low or inappropriately normal LH: the furnace could work, but nobody is asking it to. That is secondary hypogonadism.

This distinction is why LH belongs on a proper diagnostic panel rather than testosterone alone — it is the single measurement that separates the two mechanisms. The European Male Ageing Study found that axis disruptions in older men are differentially linked to age and to modifiable risk factors,[3] which is the finding with the most practical consequence on this page: the secondary pattern travels with obesity, illness and medication far more than with birthdays.

That is also the territory of functional hypogonadism — low testosterone driven by metabolic state rather than by a failing gonad, and sometimes reversible by addressing the cause.

Failure three: more of it gets bound

The third change is not production at all. Sex hormone-binding globulin rises with age, and SHBG-bound testosterone is not available to tissue.

The Massachusetts Male Aging Study measured SHBG increasing cross-sectionally at about 1.6% per year, while total testosterone declined at about 0.8% per year and free and albumin-bound testosterone declined at about 2% per year.[2] Free testosterone falls roughly twice as fast as total, because the same rising SHBG that sequesters hormone is also propping up the total figure.

A man can therefore lose a substantial fraction of his usable testosterone while his total sits in the reference range. This is the mechanism behind free versus total testosterone and what SHBG does to your results, and it is the most common reason a symptomatic man is told his labs are normal.

The methodological finding worth knowing

MMAS followed 1,156 men from an original 1,709 and compared two ways of measuring the same decline. The longitudinal decline within individual men was considerably steeper than the cross-sectional trend across men at a single point — 1.6% per year for total testosterone longitudinally, against 0.8% cross-sectionally.[2]

That is a paradox worth sitting with: following the same man over a decade shows a faster fall than comparing 50-year-olds against 70-year-olds today. The authors' proposed explanation is that incident poor health accelerates the decline, and cross-sectional sampling under-represents it.

The same study found that apparent good health — no chronic illness, no prescription medication, no obesity, no excessive drinking — added 10–15% to the level of several androgens.[2] Health status shifts the whole curve upward without changing its slope.

What counts as ageing, and what does not

BLSA is the study that isolates the age effect most cleanly. Measuring testosterone and SHBG in 890 men with a mixed-effects model, it found significant, independent, age-invariant longitudinal effects of age on both testosterone and the free testosterone index, after compensating for sampling-date artefacts. Total testosterone also decreased with increasing body mass index.[1]

So some of the decline is genuinely age. A meaningful share of it is weight, illness and medication — which is the actionable part.

BLSA also quantified how often the decline crosses into hypogonadal territory: using total testosterone criteria, roughly 20% of men over 60, 30% over 70, and 50% over 80 had hypogonadal levels, with higher proportions on free testosterone index criteria.[1]

Why the mechanism changes the decision

• Low T with high LH points at the testis. This is the least reversible pattern and the one where replacement is most straightforwardly the answer.
• Low T with low or normal LH points upstream, and often at something addressable — weight, sleep apnoea, opioids, glucocorticoids, illness. Treating the cause can raise testosterone without replacing it.
• Normal total T with high SHBG and low free T is not a production failure at all. It is an availability problem, and it is invisible on the test most men are given.

Three different biologies, three different conversations, and one lab value that cannot tell them apart on its own. That is the argument for measuring the right things at the outset.

Bottom line

Testosterone falls with age because the testes respond less, the pituitary signal weakens, and SHBG rises to bind more of what is left — three changes running simultaneously and at different rates in different men. Total testosterone, measured alone, hides all three. LH separates testicular failure from a weakened signal, and SHBG with free testosterone reveals how much of the total is actually available.

References

3 sources

1. Harman SM, Metter EJ, Tobin JD, et al. Longitudinal effects of aging on serum total and free testosterone levels in healthy men. Baltimore Longitudinal Study of Aging. J Clin Endocrinol Metab. 2001;86(2):724–31. doi:10.1210/jcem.86.2.7219 · PMID 11158037
2. Feldman HA, Longcope C, Derby CA, et al. Age trends in the level of serum testosterone and other hormones in middle-aged men: longitudinal results from the Massachusetts male aging study. J Clin Endocrinol Metab. 2002;87(2):589–98. doi:10.1210/jcem.87.2.8201 · PMID 11836290
3. Wu FC, Tajar A, Pye SR, et al. Hypothalamic-pituitary-testicular axis disruptions in older men are differentially linked to age and modifiable risk factors: the European Male Aging Study. J Clin Endocrinol Metab. 2008;93(7):2737–2745. doi:10.1210/jc.2007-1972 · PMID 18270261