Testosterone Dosing for Women: Forms, Starting Doses, and How Long Each Takes to Work

Testosterone dosing for women in specifics: cream, gel, pellets and injections compared, realistic starting doses, what to test, and when to expect a change.

The starting dose most menopause specialists use for a woman is one-tenth of the male dose. That number comes straight out of the 2021 clinical practice guideline from the International Society for the Study of Women's Sexual Health, and it exists because a healthy woman carries roughly one-tenth to one-fifteenth of the testosterone a healthy man does. Everything else on this page follows from that ratio.

You're unlikely to find that sentence on a hospital website, and the reason is structural rather than sinister. There is no FDA-approved testosterone product for women in the United States, so every American prescription is off-label, and large institutions are cautious about publishing dose-level instructions for an off-label use of a controlled substance. The specifics still exist. They sit inside guideline documents that most women never read, and they're perfectly readable once someone lays them out.

I'm a medical technologist, not a physician. What follows is what the guidelines say, what the research behind them measured, and what my own experience has been. It's the map, not the prescription. The dose that belongs on your body comes from a clinician who can examine you and read your labs.

What a normal female testosterone level looks like

Before any dose makes sense, the target does.

The ISSWSH guideline cites reference values from premenopausal women with no sexual complaints: roughly 45.5 to 57.5 ng/dL in the 20s, 27.6 to 39.8 ng/dL in the 30s, and 27.0 to 38.6 ng/dL in the 40s. Healthy men run 300 to 800 ng/dL.

Two things fall out of those numbers. The first is the one-tenth rule. The second is that a woman's level has already dropped by roughly half between her twenties and her forties, which is why so many women feel a change long before anything about their cycle announces itself.

The goal of therapy is to put you back inside the premenopausal range, not above it. The guideline is explicit that treatment should not target concentrations exceeding those of premenopausal women, and that there is no magic number to aim for, because blood levels don't predict who feels better. Levels get measured to keep you from overshooting, not to hit a bullseye.

The forms, compared honestly

Form	Typical starting point	When it works	Reversible
1% cream (AndroFeme)	5 mg/day (0.5 mL), can titrate to 10 mg/day	4 to 12 weeks	Yes, stop and it clears
Male 1% gel, off-label	One-tenth of a tube or packet daily, roughly 4 drops	4 to 12 weeks	Yes
Compounded cream	Varies by pharmacy	4 to 12 weeks	Yes
Subcutaneous injection	Small weekly or twice-weekly doses	Days to weeks	Yes, but each dose has to clear
Pellets	Implanted every 3 to 6 months	Weeks	No

Transdermal cream is the reference standard. AndroFeme, a 1% testosterone cream at 10 mg/mL, was approved in Australia in 2020 for postmenopausal women with low sexual desire, and it is the only product in the world designed and approved at a female dose. The recommended starting dose is 5 mg a day, which is half a milliliter, titrated up to 10 mg a day if needed. Clinicians outside Australia can register at doctordirect.com.au to prescribe it for direct delivery. That is a real route, and it's how I first got hold of testosterone myself, through Newson Health in London.

Male 1% gel at one-tenth the dose is the workaround when a female product isn't available, and the guideline endorses it directly. A man with hypogonadism uses 30 tubes or packets a month. A woman starts on one-tenth of a tube per day, which works out to about three tubes a month and roughly four drops a day at room temperature.

One practical detail from the guideline that will save you money and grief: ask for resealable tubes, not single-use packets. Packets let the gel evaporate, which concentrates whatever testosterone is left, so your dose drifts upward without you changing anything. And be ready for sticker shock at the pharmacy counter, because the minimum purchase is a man's monthly box of 30 tubes. That box will last you close to a year.

Compounded cream is where most American women end up, and the guidelines are blunt that they'd rather you didn't. ISSWSH, the 2019 Global Consensus Position Statement, and the National Academies of Sciences, Engineering, and Medicine all recommend against compounded testosterone, on the grounds that compounded products lack efficacy and safety data and can vary in actual concentration from what the label claims. Manufactured male products are inspected; compounded ones are not, in the same way.

Here's the tension nobody resolves for you. In a country with no approved female product, "don't use compounded" and "use one-tenth of a male product" are the only two options on the table, and plenty of prescribers reach for compounding because the dosing is cleaner. If you and your clinician go that way, the guideline's fallback position is that the pharmacy should meet industry standards for active pharmaceutical ingredient purity and Good Manufacturing Practice. Ask which pharmacy, and ask whether they do potency testing on batches. A compounding pharmacy that's proud of its process will tell you.

Injections are not recommended in the guidelines for women, because intramuscular dosing tends to produce peaks well above the female range. That verdict was written with conventional monthly or biweekly intramuscular dosing in mind. Small, frequent subcutaneous doses are a different pharmacokinetic animal, and they're increasingly used in practice precisely because they hold a steadier level. The honest state of play: the evidence base behind the guideline is transdermal, so a clinician following the guideline to the letter will steer you to a cream, and one who prescribes small subcutaneous doses is working ahead of the published evidence rather than against it. Both positions are defensible. Ask yours which one they hold and why.

Pellets get the firmest no in the guideline, and for reasons that have nothing to do with squeamishness. They frequently produce supraphysiologic levels, they can't be titrated, and once one is in, you can't take it back. If the dose is wrong, you wait months. The published counterargument, mostly from Rebecca Glaser's work with subcutaneous implants, is that higher levels produce better symptom relief and that heavier women need more, which is a real finding and one reason the pellet industry can point at data. It doesn't fix the irreversibility. There's also a specific risk that rarely comes up in a pellet clinic's sales conversation: testosterone pellets and endometrial risk.

Buccal troches and lozenges deserve a warning, because they're marketed as if they were transdermal. You hold them in your mouth, you swallow saliva, and a substantial amount goes through your gut. ISSWSH classifies them as oral testosterone therapy, and oral testosterone is the one route with a known problem, which is an unfavorable effect on lipids that the non-oral routes don't share.

Where to put it, and why the location is not arbitrary

For cream and gel, the guideline names the back of the calf, the upper outer thigh, or the buttock. If you shave the area, wait at least an hour. Use the same site consistently, because absorption varies by skin thickness and switching sites changes your effective dose without changing your prescription.

Then there's transference, which is the part women aren't warned about often enough. Testosterone moves from your skin to anyone who touches it. Young children and female partners are the ones who matter; the risk to a male partner is minimal. Let it dry, cover the site, and wash residual product off with soap and water. This is one reason the calf and the buttock are on the list and the inner wrist is not: they're less likely to end up against someone else's skin.

How long it takes, in weeks not vibes

This is the question I get asked most, and the guideline answers it with unusual precision, drawn from the clinical trial data:

Around 4 weeks: some women notice improvement, and the distress attached to the symptom starts dropping first.
6 to 8 weeks: the average point at which effects emerge.
About 12 weeks: maximal effect on desire and on satisfying sexual events.
5 to 6 months: distress continues to fall throughout this window.
6 months: the decision point. If there's been no clinically meaningful improvement by then, the guideline says stop and look for another explanation rather than pushing the dose higher.
6 to 12 months in, if it's working: consider a drug holiday to test whether you still need it.

That six-month line matters in both directions. It protects you from a clinic that keeps escalating your dose because you don't feel different yet, and it protects you from giving up at week five. Most of the failed trials in this field ran twelve weeks, which is roughly the point where the effect is only just fully arriving. The design problems run deeper than that, and I've written them up separately: why testosterone studies in women keep finding nothing.

What to test, and when

Before you start: total testosterone, plus SHBG. The total is not a diagnostic test, and no one should be telling you that a number qualifies or disqualifies you. It's there as a baseline, and to catch the woman whose level is already mid-range or high, in whom low testosterone probably isn't the explanation for how she feels.

Ask which assay your lab runs. Direct immunoassay, which is what most labs use by default, is unreliable at female concentrations. Mass spectrometry (LC-MS/MS) is accurate and reproducible, and even it hasn't been internationally standardized, so two labs can hand you two different numbers for the same blood. That's a large enough problem to have its own page: why your labs come back "normal".

3 to 6 weeks after starting: repeat total testosterone. The point is titration and making sure you're not applying too much.

Within 6 weeks of any dose increase: repeat again.

Every 4 to 6 months once you're stable: keep checking, because gel and cream doses drift upward over time in ordinary use.

If your level comes back supraphysiologic, even with no visible side effects, the guideline says titrate down and recheck in 2 to 3 weeks. A high number with no symptoms is still a number worth fixing.

On SHBG: sex hormone binding globulin is the protein that holds testosterone in circulation and keeps it from reaching your tissues. Oral estrogen, combined oral contraceptives, thyroid replacement and untreated hyperthyroidism all push it up, which drives free testosterone down while total testosterone can look perfectly fine. If you're on the pill and your total reads normal, that mechanism is probably why: birth control, SHBG and your free testosterone. The guideline suggests calculating free testosterone from total and SHBG when a woman's total is at the top of the range and she still isn't responding, and it points at the ISSAM free testosterone calculator for the arithmetic.

One more possibility for non-responders, and the guideline flags it as a proposal rather than an established fact: some experts think women with reduced 5-alpha reductase activity may not respond to physiologic transdermal doses at all.

Testosterone alongside estrogen and progesterone

Testosterone is almost never the first hormone you're given, and it shouldn't be. It's usually added on top of established estrogen therapy, with progesterone if you still have a uterus, once estrogen has been optimized and something is still missing. That order exists because a lot of what women attribute to low testosterone (the sleep disruption, the flatness, the joint pain) responds to getting estradiol right first.

The scale of the evidence differs sharply between the two. Estrogen therapy for menopausal symptoms is well-studied and approved. Testosterone for women has exactly one evidence-based indication in the 2019 Global Consensus Position Statement, and it is low sexual desire with associated distress in postmenopausal women. Everything else falls outside that, and the consensus is not vague about how it falls outside. On mood it reports a null result at the highest level of evidence: available data do not show an effect of testosterone on depressed mood (Level I, Grade B), and show no effect on general wellbeing (Level I, Grade A). Cognition is the outcome graded insufficient evidence, meaning the trials haven't been done rather than that they came back empty. So the mood lift I experienced runs ahead of what the trials show. I'd rather tell you that straight than sell it to you as an under-researched benefit.

The most useful recent data on that gap comes out of the UK. A 2025 retrospective cohort in the British Journal of Psychiatry followed 920 women at Britain's largest specialist menopause clinic and found mean mood scores fell about 45% after roughly 107 days on hormone therapy, with improvement in both perimenopausal and postmenopausal women, across regimens that included transdermal testosterone. Real-world data with no control arm can't separate the drug from the attention, the expectation, or the natural course, and the authors say so. It's the best available signal, not proof.

An earlier pilot from the same clinic, 510 peri- and postmenopausal women on transdermal testosterone, found the symptom with the highest response rate wasn't libido at all. It was "loss of interest in most things," at 56%. That result is why this site exists, and it has its own piece: low testosterone or depression, and how to tell.

On perimenopause specifically: the 2019 consensus indication says postmenopausal, and the reason is simply that the trials enrolled postmenopausal women. The 2021 ISSWSH guideline goes a step further and says limited data also support use in late reproductive age premenopausal women. Specialists prescribe in perimenopause routinely, and the UK cohort above included 448 perimenopausal women, so the practice is ahead of the strongest evidence. Worth knowing before someone tells you your symptoms don't count yet.

What microdosing actually means here

"Microdosing testosterone" has picked up a second meaning online, borrowed from gender-affirming care, and the two are not the same thing. When a menopause specialist says low-dose or microdose, she means a dose designed to land you inside the premenopausal female range: about 5 mg a day of 1% cream, or one-tenth of a male gel dose. That is not a small version of male therapy. It's replacement, aimed at a number your own body used to produce.

The dose that produces masculinizing changes is five to twenty times higher. Which is why the side effect list that frightens women off is mostly borrowed from a population on a completely different dose: testosterone side effects in women, and what's really just too high a dose.

My own dosing, and why I'm not the model

I started on AndroFeme cream as a patient at Newson Health in London. Follow-up bloods showed I wasn't absorbing enough of it, and my working theory is that being overweight at the time was part of it, since skin thickness and body composition genuinely change how much of a topical gets into circulation. I felt nothing. That happens, and it's the strongest argument for checking a level three to six weeks in rather than assuming the cream is working because you applied it.

I moved to an injectable. I use Despamen, a combination product sold over the counter in Mexican pharmacies: testosterone enanthate 100 mg with estradiol valerate 5 mg in one pre-filled syringe. The label says once a month. Testosterone enanthate has a half-life of about 4.5 days, so a monthly injection peaks in the first days and falls through subtherapeutic ranges by weeks two and three. I could feel that window open every month. The drive went quiet, the mornings flattened, and the anhedonia came back on a schedule I could predict.

So I inject twice a month instead. 100 mg of testosterone enanthate is a male replacement dose, and taking it twice monthly puts me above the physiologic female range. I know that. I'm knowingly taking a supraphysiologic dose, and every reassuring thing the global consensus says about side effects is conditioned on a dose I'm not taking. I made that trade with my eyes open because the alternative on offer was two weeks a month of the thing I started treatment to escape. It's my decision about my own body, and it is not a protocol to copy.

What I'd want instead is the opposite shape: smaller amounts more often, twice weekly or every other day, holding a steady level rather than a spike and a slide. That would lower my total exposure and make me feel better rather than worse. It's what I'm working on.

I put this here because the gap between the guideline and what a real woman ends up doing is the most useful thing I can show you, and pretending my own dosing is textbook wouldn't help anyone.

Questions I get asked

What's the best testosterone for women? The cream, if you can get it and you absorb it. AndroFeme is the only product in the world formulated and approved at a female dose, and transdermal delivery gives you the steadiest levels with the easiest way to change your mind. Male 1% gel at one-tenth the dose is the practical substitute where AndroFeme isn't available.

What's the testosterone gel dosage for a female? One-tenth of the male starting dose. With 1% gel in resealable tubes, that's roughly four drops a day, about one-tenth of a tube, or three tubes a month.

How long does testosterone cream take to work? Some women notice something at four weeks. The average is six to eight. Peak effect on desire lands around twelve weeks, and improvement in distress keeps building for five or six months. If nothing has changed by six months, the guideline says look for a different cause rather than raising the dose.

Can you take testosterone in perimenopause? Specialists do it regularly, and the largest real-world cohort included hundreds of perimenopausal women. The 2019 global consensus indication is postmenopausal because that's who the trials enrolled, while the 2021 ISSWSH guideline says limited data also support late reproductive age premenopausal women. You're in a gap between the evidence and the practice, which is worth naming out loud with your prescriber rather than discovering later.

Is testosterone safe after menopause? At doses that keep you inside the premenopausal range, the global consensus found no association with hair loss, clitoral enlargement or voice change, and non-oral routes don't carry the lipid problem that oral testosterone does. The honest limit is duration: the pooled safety data don't extend much beyond two years, so nobody can tell you what twenty years looks like.

Do I need to be on estrogen first? Usually, yes, and most specialists optimize estradiol (with progesterone if you have a uterus) before adding testosterone. Testosterone therapy has been studied both with and without concurrent estrogen, so it isn't a hard requirement, but sequencing it second means you find out what estrogen alone can fix.

None of this replaces a clinician. What it does is let you walk into the appointment knowing what a starting dose looks like, when to expect a change, and which test to ask for, so the conversation starts somewhere useful. If you're not sure you're a candidate at all, start with do I need testosterone?, which includes the part where the answer is often no. If you have the symptoms and no one to write the prescription, who will prescribe testosterone to a woman is the next page.

I'm a medical technologist sharing my own reading of the research and my own experience, not your doctor. Take this to someone who can examine you.

The other pillars on this site: testosterone, libido and desire in perimenopause · perimenopause symptoms and irregular periods · menopause brain fog

Sources

Dosing, monitoring and timelines - Parish SJ, Simon JA, Davis SR et al., Journal of Women's Health 2021, the ISSWSH Clinical Practice Guideline for the Use of Systemic Testosterone for Hypoactive Sexual Desire Disorder in Women. Source for the one-tenth-of-a-male-dose rule, the four-drops-a-day figure, the AndroFeme 5 mg starting dose, the application sites and transference guidance, the female reference values by decade, the 4 / 6-8 / 12-week and 6-month timelines, the 3-to-6-week and 4-to-6-month monitoring schedule, the assay limitations, and the position on compounded products, pellets, injections and buccal troches. - Davis SR, Baber R, Panay N et al., J Clin Endocrinol Metab 2019, the Global Consensus Position Statement on the Use of Testosterone Therapy for Women, endorsed by eleven societies: the only evidence-based indication is hypoactive sexual desire disorder in postmenopausal women, available data show no effect on depressed mood (Level I, Grade B) or on general wellbeing (Level I, Grade A), and cognitive outcomes are the ones classified as insufficiently evidenced. - Islam RM, Bell RJ, Green S, Page MJ, Davis SR, Lancet Diabetes and Endocrinology 2019, the meta-analysis of randomized trial data underpinning the consensus: benefit for sexual function, no adverse lipid effect from non-oral physiologic dosing, and safety data that do not extend beyond 24 months.

Mood and cognitive outcomes - Glynne S, Kamal A, McColl L et al., British Journal of Psychiatry 2025: retrospective cohort of 920 women (448 perimenopausal, 435 postmenopausal) at the UK's largest specialist menopause clinic; mean Meno-D scores fell 44.59% after an average of 107 days. Real-world and uncontrolled. - Glynne S, Kamal A, Kamel AM, Reisel D, Newson L, Archives of Women's Mental Health 2024: pilot cohort of 510 peri- and postmenopausal women on transdermal testosterone; "loss of interest in most things" had the highest response rate at 56%. Observational, no placebo arm.

The higher-dose argument, presented for completeness - Testosterone Therapy in Women: A Literature Review of Clinical Indications, Dosing, Routes of Administration, and Safety, Cureus 2026: reviews the case that theoretical physiologic doses may be clinically insufficient and that subcutaneous implant doses of 55 to 240 mg have been used since 1938. This review reads the pellet literature more favorably than the global consensus does; both positions are here so you can see the disagreement rather than be handed one side of it.