Do Women Ever Need Testosterone?

Testosterone is often labelled a “male hormone”, yet women produce it throughout life and rely on it for multiple aspects of health. Produced in the ovaries and the adrenal glands, testosterone in women supports sexual desire, energy, mood, muscle maintenance, bone density and overall vitality. Circulating levels are roughly an order of magnitude lower than in men, but the hormone remains biologically meaningful. So the question of whether women ever need testosterone comes down to understanding individual hormone balance, recognising when levels and symptoms genuinely align, and approaching any intervention with careful medical oversight.

Understanding hormones in women

Women's endocrine systems operate through an interplay of oestrogen, progesterone, testosterone and other signals. Oestradiol, the main oestrogen of the reproductive years, drives menstrual cycling and is central to bone protection. Progesterone opposes oestrogen's effect on the endometrium — the clearest demonstration being the PEPI trial, in which 27.7% of postmenopausal women given oestrogen alone developed simple endometrial hyperplasia against 0.8% on placebo, a risk that adding a progestogen removed. Testosterone, though present in far smaller amounts, contributes to libido, arousal, motivation and lean tissue.

These hormones do not act in isolation. Shifts in one influence the others, and the most clinically useful example is sex hormone-binding globulin, or SHBG — the protein that binds most circulating testosterone. Because only the unbound fraction is available to tissues, anything that moves SHBG moves the amount of usable hormone without changing total testosterone at all. Oral oestrogen raises SHBG and therefore lowers free testosterone; that is a real effect of a common prescription, not a quirk of the assay.

When testosterone falls relative to a woman's own baseline, symptoms may appear even while absolute numbers sit inside a broad reference range. Reduced sexual desire or satisfaction, unexplained fatigue, diminished motivation, mood flattening and loss of muscle tone are all commonly reported. They also overlap with a great many other conditions, from thyroid disease to iron deficiency to simple sleep debt, so hormone status is one piece of a much larger picture rather than the explanation by default.

What the blood tests actually measure

Accurate assessment begins with properly timed blood work interpreted by a clinician experienced in women's hormone health. Total testosterone, free testosterone and SHBG provide the core androgen picture. In Australia these are reported in nmol/L, not the ng/dL figures that circulate in American articles — the conversion is 1 ng/dL = 0.0347 nmol/L.

The real numbers are lower than most people expect. In 588 healthy Australian women aged 18 to 39 with regular cycles and no polycystic ovary syndrome, measured by liquid chromatography–tandem mass spectrometry, median total testosterone across the cycle was 0.34 nmol/L (about 9.8 ng/dL), with individual values spanning 0.04 to 1.01 nmol/L (roughly 1.2 to 29 ng/dL). Pooled mass-spectrometry data put the premenopausal female reference interval at about 0 to 1.7 nmol/L, against 7.7 to 29.4 nmol/L for men aged 18 to 40 — a difference the same authors describe both as an order of magnitude and, for young adults, as roughly 15-fold.

Two cautions follow from this, and both matter more than any single number. First, the assay method is not a technicality. Immunoassays were optimised for male concentrations and become progressively less accurate at the much lower levels found in women, which is why mass spectrometry is the reference method and why the international consensus statement says direct assays are appropriate in clinical practice if mass spectrometry is unavailable, to exclude a high baseline level and to exclude supraphysiological levels during treatment. Second, a single range is the wrong construct: in that Australian cohort every C19 androgen was significantly lower at ages 35 to 39 than at 18 to 25, and the authors concluded that age-specific reference ranges are required. Absolute values from one mass-spectrometry service also cannot be transplanted onto another, because these assays are not harmonised between laboratories.

Oestradiol is usually measured alongside. It fluctuates widely across the cycle — low in the early follicular phase, peaking before ovulation, then falling and rising again to a second, smaller luteal peak — and is very low after menopause, under about 26 pmol/L (7 pg/mL) by reference-method mass spectrometry, though the cut-off a laboratory quotes depends heavily on the method it runs. FSH and LH are worth understanding for what they are: tests of ovarian and pituitary function used for menopause staging and fertility assessment. They are not tests of androgen status, and they do not appear anywhere in the biochemical criteria for hyperandrogenism in the current international PCOS guideline. For androgens, the timing advice is a morning sample, drawn in the follicular phase in women who are still cycling, with repeat testing when a result is borderline.

How age plays a role — and the part most articles get wrong

Testosterone in women is highest in early adulthood and declines with age, and that decline begins well before menopause. The Australian data above show it inside the reproductive years alone. In 1,104 Australian women aged 40 to 69, median testosterone fell about 25% from age 40–44 to 55–59 (0.56 to 0.42 nmol/L), reached its lowest point at 58–59, and then recovered.

Here is where the popular account goes wrong. It is routinely written that ovarian testosterone production “largely ceases” at menopause, with adrenal output taking over. The longitudinal evidence says otherwise, and the clearest dataset is Australian. In the Melbourne Women's Midlife Health Project, 172 women were sampled annually through natural menopause across seven years. The finding, verbatim: “Mean T levels did not vary with time relative to FMP and were independent of age and BMI”, and the conclusion, “T remains unchanged during the menopausal years.”

What changes at menopause is not testosterone but its carrier. In that same cohort SHBG fell by 43% from four years before to two years after the final period. Because the free androgen index is testosterone divided by SHBG, it rose by 80% over that window. That index is a calculated ratio rather than a measured bioavailable fraction, so it should not be over-read — but the direction is the opposite of the usual story. Across the menopause transition itself, oestradiol falls steeply while testosterone holds steady and the unbound proportion of it goes up.

Two things follow. The testosterone decline women experience is age-related, not menopause-related, which is why it starts in the thirties and does not accelerate at the final period. And the postmenopausal ovary is not a spent organ: it continues to secrete androgen, which is why removing the ovaries is a different event altogether. After risk-reducing removal of both ovaries, testosterone, androstenedione, oestradiol and oestrone all fall significantly within six weeks and remain lower at seven months. Surgical menopause genuinely does cause an abrupt androgen drop. Natural menopause does not.

Later life adds one more wrinkle. In 5,326 Australian women aged 70 and over, testosterone and oestrone increased with age while adrenal DHEA continued to fall — so the late-life rise cannot be explained by the adrenal glands picking up the slack, as is often claimed.

Peptides that act upstream: what the evidence in women actually shows

Some peptides influence hormone production indirectly by acting on upstream signalling rather than supplying a hormone. Kisspeptin is the most studied. It acts on KISS1R receptors on GnRH neurons in the hypothalamus, increasing GnRH pulses, which drive pituitary release of LH and FSH; LH then acts on the ovary. That mechanism is well established.

What the human data in women show, though, is something more specific than a general hormone boost: the response depends enormously on where a woman is in her cycle. In eight healthy women given the same dose of kisspeptin-54 in each cycle phase, the mean rise in LH was 0.12 IU/L in the follicular phase, 2.17 IU/L in the luteal phase and 20.64 IU/L in the preovulatory phase. That is not a modest phase effect; it is the difference between almost nothing and a very large response, from an identical injection. Postmenopausal women, meanwhile, produce the largest gonadotropin response of any female group to a single bolus, yet are described in the literature as “resistant to the stimulatory effect of continuous kisspeptin” on LH secretion.

On testosterone specifically, the honest position is that the evidence in women is thin. The often-quoted figure — testosterone rising from 16.6 to 24.0 nmol/L on a 22.5-hour infusion — comes from four men, and male numbers in that range are not transferable to female physiology. In women, androgens have been measured, with mixed results: a randomised, placebo-controlled crossover trial in 32 premenopausal women with hypoactive sexual desire disorder included serum testosterone as a pre-specified secondary outcome and found LH rose by a mean of 2.14 IU/L; beyond that, the androgen data amount to small exploratory work. Kisspeptin's real clinical research programme in women is about fertility and sexual brain processing — triggering oocyte maturation in IVF, rescuing ovulation in PCOS, and modulating sexual and emotional responses in HSDD — not androgen replacement.

Growth hormone peptides: read the participant lists

Tesamorelin, CJC-1295 and ipamorelin raise growth hormone and IGF-1, and they are frequently discussed as an indirect route to better hormonal health. Three specifics are worth knowing before drawing conclusions about women.

First, the human evidence base for ipamorelin is a single 1999 dose-escalation study in 40 healthy men, eight per dose level, which showed one short GH pulse peaking at about 40 minutes. There is no published human trial of ipamorelin in women. Second, tesamorelin has the strongest body-composition evidence of the three, and one trial is far more relevant to women than the commonly cited HIV studies: a 12-month randomised, placebo-controlled trial in 60 abdominally obese adults with reduced GH secretion — 21 of them women, all with FSH at or below 20 IU/L and so not postmenopausal — found a visceral fat treatment effect of −35 cm² (95% CI −58 to −12, P = 0.003) alongside falls in carotid intima-media thickness, C-reactive protein and triglycerides. The larger HIV datasets — a pooled phase 3 analysis showing a −15.4% visceral fat treatment effect at 26 weeks in 806 patients, and a JAMA trial showing a mean 34 cm² reduction over six months — were conducted in predominantly male HIV-positive populations.

Third, the sleep claim does not survive testing. The one substantial randomised trial to measure sleep prospectively gave tesamorelin or placebo to 152 adults aged 55 to 87 for 20 weeks and assessed it with the Pittsburgh Sleep Quality Index. There was no effect on sleep (P = 0.98) and none on mood (P = 0.31). Cognition improved; sleep did not. Anyone told that these peptides reliably improve sleep is being told something the best available trial specifically did not find.

One more argument deserves correcting because it is imported wholesale from men's articles. In men, adipose aromatisation of androgens to oestrogen is framed as a problem to be minimised. In women it is not the same thing at all: adipose tissue is the principal site of oestrogen formation in postmenopausal women, and the fractional conversion of androstenedione to oestrone rises with both age and body weight. Reducing visceral fat has real cardiometabolic benefits, well supported by the trials above. But “less aromatisation” is not a straightforward win in a woman who no longer has ovarian oestrogen production, and presenting it as one misreads the physiology. The genuine growth-hormone interaction worth knowing is that the route of oestrogen therapy matters: oral and transdermal oestrogen have contrasting effects on 24-hour GH secretion and IGF-1, a finding from the Garvan Institute in Sydney.

Where testosterone therapy for women is actually approved

The governing guidance is the Global Consensus Position Statement on the Use of Testosterone Therapy for Women (Davis SR and colleagues, 2019), published simultaneously in four journals and endorsed by the International Menopause Society, The Endocrine Society, and in Australia by the Endocrine Society of Australia and RANZCOG. Its position is narrow and worth quoting accurately. It supports a trial of testosterone for postmenopausal women with hypoactive sexual desire disorder, at doses producing concentrations that do not exceed premenopausal physiological levels, after other contributors have been addressed.

For everything else, the statement is more definite than “limited evidence”. On mood: “Available data do not show an effect of testosterone on depressed mood” — Level I evidence. On general wellbeing: “Available data show no effect of testosterone therapy on general wellbeing” — Level I, Grade A. That is not an evidence gap awaiting better studies; it is good evidence of no effect, and it is a meaningful distinction for anyone hoping testosterone will fix fatigue.

On availability, the widely repeated line that no testosterone product is approved for women is wrong in Australia, and Australia is unusual here. AndroFeme 1, a testosterone 1% cream made by Lawley Pharmaceuticals in Perth, was TGA-registered on 23 November 2020 for the treatment of hypoactive sexual desire dysfunction in postmenopausal women. New Zealand's Medsafe has approved the same product. The only earlier approval anywhere, the Intrinsa patch authorised in Europe in 2006 for surgically menopausal women, was later withdrawn from the market. So an Australian woman discussing this with her doctor is discussing a locally registered medicine with an approved indication, not an off-label improvisation — which is a materially different conversation from the one described in most American articles.

The cost caveat is real, though. At its November 2025 meeting the Pharmaceutical Benefits Advisory Committee did not recommend listing testosterone 1% cream for HSDD in postmenopausal women, so it remains a private prescription rather than a PBS-subsidised one.

Always consult a doctor and specialist

Hormone evaluation and any form of support — lifestyle, conventional hormone therapy or peptides — should be guided by a qualified physician, ideally one with expertise in women's endocrine or menopause care. Self-directed testing risks missing an underlying condition, misreading a method-dependent number, or overlooking simpler foundational approaches. A thorough assessment usually includes symptom history, examination, targeted blood work, and a discussion of reproductive goals, cardiovascular risk, bone health and personal preference.

In short: women do produce testosterone, it does matter, and it is reasonable to have it assessed — particularly where low desire is causing distress. But the decline is age-related rather than a menopausal cliff, the numbers mean little without knowing the assay and the woman's age, the peptide evidence in women is far thinner than the marketing suggests, and in Australia there is an approved, regulated option to discuss. The path forward starts with professional evaluation and shared decision-making, not an isolated number.