Testosterone in women: useful, narrow, and oversold

This is a site about men. This page exists anyway, for three reasons.

The first is that men ask. A partner has been offered testosterone, or has read that it will fix her energy, and the question lands in a men’s appointment.

The second is that the marketing pattern is identical. The same arguments used to sell testosterone to men who don’t need it — that a symptom means a deficiency, that a number means a diagnosis, that more is better — are being used on women, with a thinner evidence base and no approved product. A man who has learned to recognise that pattern in his own care can recognise it in someone else’s.

The third is specific and practical. A man using a topical testosterone gel can transfer testosterone to the people he touches. That is a labelled warning, not a theoretical concern, and it is the one part of this subject that genuinely belongs on a men’s site rather than a women’s one.

For the clinical detail — dosing, monitoring, how it fits alongside estrogen and progesterone, and how it is prescribed in practice — the full chapter is on our sister site: Testosterone therapy for women, at MenoExperts.

Women make testosterone, and it is not a trace hormone

Testosterone in women comes from the ovaries and the adrenal glands, and circulates at roughly a tenth to a twentieth of male concentrations. It is not a vestigial male hormone that women happen to carry; it is a functioning part of female endocrine physiology, and it is also the substrate from which some estradiol is made.

Two things about the trajectory are worth stating plainly, because both are commonly misdescribed.

Testosterone in women declines gradually across the reproductive years, beginning well before menopause. It does not fall off a cliff at menopause the way estradiol does. The background to the 2019 international consensus statement makes this point explicitly, and it matters: “menopause” is not the event that explains a woman’s testosterone level, so a low level is not automatically a menopausal finding.

The exception is surgical menopause. Removal of both ovaries produces an abrupt fall in circulating testosterone, and that is a different clinical situation from natural menopause.

The one indication with good evidence behind it

Hypoactive sexual desire disorder in postmenopausal women — persistently low sexual desire that causes the woman herself distress.

This is the only indication endorsed by the 2019 Global Consensus Position Statement on the Use of Testosterone Therapy for Women, led by Susan Davis and published simultaneously in Climacteric, Maturitas, the Journal of Sexual Medicine and the Journal of Clinical Endocrinology and Metabolism. It carries an unusually broad set of endorsements — the International Menopause Society, the Endocrine Society, the European Menopause and Andropause Society, the International Society for Sexual Medicine, ISSWSH, the North American Menopause Society, the Royal College of Obstetricians and Gynaecologists and others. When that many societies sign the same document, the disagreement that characterises so much of the testosterone field is largely absent here.

The evidence underneath it is a systematic review and meta-analysis by Rakibul Islam and colleagues in The Lancet Diabetes & Endocrinology in 2019: 36 randomised controlled trials, 8,480 women. In postmenopausal women, testosterone significantly improved satisfactory sexual event frequency (mean difference 0.85, 95% CI 0.52–1.18), sexual desire (standardised mean difference 0.36, 95% CI 0.22–0.50), arousal, orgasm, pleasure, responsiveness and self-image, and reduced sexual distress (SMD −0.27, 95% CI −0.36 to −0.17).

Read that first number honestly. A mean difference of 0.85 satisfactory sexual events is roughly one additional event per month over placebo. That is a real effect, reproducible across a large number of trials, and it is also a modest one. Some women find it clearly worth having. Some do not. Both responses are reasonable, and a woman deciding is entitled to the actual magnitude rather than the adjective.

Evidence: Established — for this indication, in this population, at physiologic doses.

What it is not established for

Testosterone is marketed to women for energy, mood, brain fog, muscle, bone, weight and ageing. The trial evidence does not currently support those uses.

The same Lancet meta-analysis reported no effect of testosterone on body composition, musculoskeletal variables or cognitive measures — with the honest caveat, which the authors state themselves, that relatively few women contributed data to those outcomes. That is a different statement from “it doesn’t work,” and we should not overstate it in the other direction. It means the studies that would justify prescribing for those reasons have not been done, and the ones that exist are null.

Route matters for one measurable harm: oral testosterone raised LDL cholesterol and lowered HDL and triglycerides; non-oral routes — transdermal patch or cream — did not. This is one reason transdermal delivery is the recommended route rather than a stylistic preference.

Evidence: Limited to Unsupported, depending on the specific claim. Fatigue and “low energy” in a woman deserve the same workup they deserve in a man: thyroid, iron, sleep, depression, medication, and everything else on the list, before a hormone is assumed to be the cause.

There is no FDA-approved testosterone product for women

Every prescription of testosterone to a woman in the United States is off-label. There is no approved female formulation. This is not a fringe observation; it is stated in the consensus documents themselves.

What follows from that is practical. The ISSWSH clinical practice guideline (Parish, Simon, Davis and colleagues, Journal of Sexual Medicine, 2021) sets out how to prescribe within those constraints: a government-approved male transdermal formulation, used cautiously at approximately a tenth of the male dose, with total testosterone monitored to keep concentrations in the premenopausal physiologic range — not above it.

The same guideline is explicit that compounded products cannot be recommended, on the grounds of absent efficacy and safety data. Pellets deserve their own sentence: once implanted, the dose cannot be adjusted, and levels are frequently supraphysiologic for months. A woman who develops androgenic side effects from a pellet has no way to stop the exposure.

A number does not make the diagnosis

The ISSWSH guideline states that a total testosterone level should not be used to diagnose hypoactive sexual desire disorder. It is measured as a baseline, so that treatment can be monitored and supraphysiologic exposure avoided — not to establish that a problem exists.

That should sound familiar. It is the same principle this site applies to men, arrived at independently by a different set of societies looking at a different sex: the symptom and its context make the diagnosis, and the laboratory value tells you what to do next and whether you have overshot.

Safety: what is known and where the data stop

In the randomised trials, testosterone made acne and unwanted hair growth significantly more likely, and produced a small overall increase in weight. No serious adverse events were recorded. Voice deepening and clitoral enlargement are the side effects women most often ask about; these are associated with supraphysiologic exposure rather than physiologic dosing, and are the reason levels are monitored.

Where the data stop matters more than where they run. Long-term safety has not been established. Breast cancer outcomes and cardiovascular outcomes over years of use have not been studied in trials designed to answer those questions. Anyone who tells a woman that testosterone is proven safe over a decade is describing evidence that does not exist — in either direction.

Evidence: Established for short-term physiologic dosing; absent for long-term outcomes.

If you use a testosterone gel, this concerns your household

Topical testosterone gels carry a boxed warning for secondary exposure. Virilization has been reported in children who came into contact with treated skin, and the FDA required the warning after those reports. Women can be exposed by the same route.

This is not a reason to avoid gel. It is a reason to use it as labelled:

• Apply to the sites specified on your product’s label, and cover the area with clothing once dry.
• Wash your hands thoroughly with soap and water immediately after applying.
• Wash the application site with soap and water before any skin-to-skin contact — this specifically includes a partner and, obviously, children.
• If a partner develops new acne, unusual hair growth or menstrual changes while you are using gel, raise transfer as the first hypothesis rather than the last.

Injectable and pellet formulations do not carry this risk. If transfer cannot be managed reliably in your household, that is a legitimate reason to discuss changing formulation — see treatment.

Where to read further

MenoExperts covers this properly, in the context it belongs in — alongside estrogen, progesterone, and the rest of a woman’s menopause care rather than as an isolated hormone.

Testosterone therapy for women →

Questions patients ask