Key takeaways
- DHEA is largely substrate rather than an active hormone — target tissues convert it locally into androgens and estrogens using their own enzymes.
- Because much of the conversion happens inside tissues, supplementation does not produce a predictable rise in serum testosterone.
- DHEA-S falls roughly 1-2% per year from the mid-twenties, a steeper decline than testosterone shows.
- Replacement clearly helps in adrenal insufficiency; a two-year randomised trial in age-related decline found no meaningful benefit.
- If it is used, test before starting, titrate against repeat DHEA-S, and monitor estradiol and PSA — what it converts into is the point.
DHEA occupies an awkward position. It is the most abundant circulating steroid in the human body, it falls further with age than almost any other hormone, and it is available over the counter in the United States without a prescription — which is exactly the combination that produces overclaiming. The honest summary is narrower than the marketing and wider than the dismissal: DHEA is a precursor with a real, well-characterised biology, a specific population in whom replacement clearly helps, and a much larger population in whom carefully conducted trials have found very little. Knowing which group you are in is the entire question.
What DHEA is, and where it does its work
Dehydroepiandrosterone is produced mainly by the zona reticularis of the adrenal cortex, with a small testicular contribution. Most of it circulates as DHEA-S, the sulphated storage form, which has a far longer half-life and is what laboratories measure.
The important mechanistic point is that DHEA is not, in the main, a hormone acting on its own receptor. It is substrate. Peripheral tissues take it up and convert it locally into testosterone, dihydrotestosterone and estradiol using enzymes those tissues express — a process sometimes called intracrinology, because the active hormone is made and used inside the same cell rather than being delivered by the bloodstream. Skin, bone, brain, adipose and genital tissue all do this to different extents.
Two consequences follow. First, DHEA's effect depends on the enzymatic machinery of the recipient tissue, which varies between individuals and explains a good deal of the variability in response. Second, giving DHEA does not produce a predictable rise in serum testosterone in men the way an androgen does — much of the conversion happens locally and never appears in a systemic measurement. The precursor cascade follows those pathways in more detail.
The decline, which is genuinely steep
DHEA-S peaks in the mid-twenties and falls steadily thereafter, roughly 1-2% per year, until it sits at around a fifth of its peak by the eighth decade (Orentreich et al., J Clin Endocrinol Metab 1984). That fall is steeper than the age-related decline in testosterone, and it is reliable enough that DHEA-S has been proposed as a marker of biological ageing in its own right.
The decline correlates with lower energy, libido, mood resilience and immune function. Correlation is doing a lot of work in that sentence — the same decades bring falling activity, worsening sleep, accumulating visceral fat and rising medication burden, all of which produce the same complaints. This is precisely why the interventional evidence matters more than the association.
Testing, and how to read the result
Order DHEA-S, not DHEA. Free DHEA fluctuates through the day; the sulphated form is stable enough to be interpretable from a single morning draw. The DHEA-S marker article covers assay behaviour and the age-referencing problem.
| Age range | Lab "normal" (µg/dL) | Where clinicians tend to aim |
|---|---|---|
| 30s | 120-520 | Upper third of range |
| 40s | 95-530 | 250-400 |
| 50s | 70-500 | 200-350 |
| 60s | 40-325 | 150-250 |
| 70+ | 40-300 | 120-200 |
Note how wide the reference ranges are. A man in his fifties can sit at 80 or at 480 and be reported as normal on both. That is the usual argument for reading position within the range rather than in-or-out — but it is also where enthusiasm outruns evidence, because being in the lower part of a wide range is not by itself a demonstration that raising it will change anything.
What supplementation has actually shown
This is the section that should determine the decision, and it points in two directions depending on the population.
Where it clearly works: adrenal insufficiency. In people whose adrenal cortex is not producing DHEA at all, replacement improves wellbeing, mood and sexual function against placebo (Arlt et al., N Engl J Med 1999). This is replacement of an absent hormone, and it behaves like replacement usually does.
Where it largely does not: age-related decline in otherwise healthy adults. A two-year randomised, placebo-controlled trial of DHEA in older adults with low DHEA-S — the exact population the supplement is marketed to — found no meaningful benefit in body composition, physical performance, insulin sensitivity or quality of life (Nair et al., N Engl J Med 2006).
That is a well-designed, adequately long, negative trial, and it should carry more weight than it usually does in discussions of this molecule. It does not prove DHEA never helps anyone. It does mean that anyone recommending it for age-related decline is working against the best available randomised evidence, and should say so rather than imply the opposite.
Who it might still make sense for
Given the above, the reasonable indications are narrow.
- Documented adrenal insufficiency, where this is replacement rather than optimisation.
- A markedly low DHEA-S for age — bottom of the range rather than merely below the midpoint — alongside symptoms that have survived a proper workup.
- Men already on testosterone therapy whose DHEA-S remains low, where the question is whether the adrenal contribution is worth restoring alongside the gonadal one. The full treatment guide covers how that is assessed.
- A trial with a defined endpoint and a stop date, rather than an open-ended addition to a stack.
It is not appropriate in men with prostate cancer, and it is a poor first move in a younger man with low testosterone and low-testosterone symptoms. In that group the yield is much higher from the things that actually suppress the axis: sleep, weight, alcohol, chronic stress, overtraining, medications. The low-testosterone checklist works through those, and the 60-second assessment is where evaluation starts.
If it is used, how it should be run
Three principles matter more than any particular regimen, and the regimen itself is a prescribing decision made after evaluation rather than something to copy.
Test before, not after. Supplementing without a baseline DHEA-S makes the follow-up uninterpretable, and a normal baseline is the commonest reason there is nothing to gain.
Start low and titrate against repeat labs, aiming for the upper-middle of the age-referenced range rather than the top of the young-adult range. Retesting at around 8-12 weeks is the usual interval, because that is long enough for a steady state and short enough to catch an overshoot.
Watch what it converts into, not just what you took. Because DHEA is substrate for both androgens and estrogens, the markers worth following are estradiol, total and free testosterone, PSA in older men, and a lipid panel. Estradiol management and PSA monitoring apply here for the same reasons they apply on testosterone therapy.
Product quality is a real issue with an over-the-counter steroid precursor. Content has been shown to vary considerably between products, which is an argument for pharmaceutical-grade or third-party-verified sourcing rather than the cheapest bottle available.
What can go wrong
- Estradiol elevation through aromatisation — water retention, breast tenderness, mood change
- Acne and oily skin, usually early and dose-related
- Accelerated hair loss in men with the genetic predisposition, via the DHT pathway
- Prostate effects — uncertain rather than established, which is itself a reason to monitor in older men
- Lipid changes, occasionally unfavourable
- Mood change, usually positive but not reliably so
The clinical pearl: DHEA is a small, targeted lever with one clear indication and a well-conducted negative trial in the population it is mostly sold to. It is not a testosterone substitute, and it will not compensate for the sleep, weight and training inputs underneath a low androgen picture. Test first, treat a documented deficiency rather than a number in the lower half of a wide range, and set an endpoint before you start.
Bottom line
DHEA is a precursor rather than an active hormone in its own right, converted to androgens and estrogens inside the tissues that use them, and it declines steeply with age. Replacement is clearly worthwhile in adrenal insufficiency. In healthy older adults with an age-typical decline, a two-year randomised trial found essentially nothing — which is the single most useful fact on this page. Where it is used, it should follow a documented low DHEA-S, be titrated against repeat testing, and be monitored for what it converts into. Expectations should be modest, and anyone promising transformation is selling rather than advising.
Educational content, not medical advice. Laboratory interpretation and any treatment decision are made by a licensed physician after individual evaluation. Individual results vary.
