Testosterone for Women: What the Evidence Actually Supports

The short answer

The same meta-analysis and the same consensus statement are equally clear about what testosterone did not do. There was no demonstrated benefit for mood, wellbeing, fatigue, cognitive performance, bone mineral density, or body composition at the doses studied. The consensus statement explicitly recommends against prescribing testosterone for those purposes. If a clinic is offering testosterone for energy, brain fog, muscle tone, or "optimization," it is prescribing outside the evidence base, and the honest framing for that is experimental, not restorative.

The practical complication in the United States is that no testosterone product is approved by the FDA for use in women. That single regulatory fact drives most of the risk in this area: it pushes prescribing toward male products dosed down by eye, or toward compounded creams and pellets that fall outside FDA manufacturing oversight. Both the National Academies and ACOG have raised concerns about compounded hormone preparations, and guidelines specifically advise against formulations — pellets and injections in particular — that produce blood levels above the normal premenopausal range. This page covers what the evidence supports, how HSDD is actually diagnosed, what dosing and monitoring should look like, what side effects are dose-dependent versus potentially permanent, and what remains genuinely unknown.

What the trials actually measured, and what they found

The 2019 systematic review and meta-analysis in Lancet Diabetes & Endocrinology pooled randomized, placebo-controlled trials of testosterone in women. In postmenopausal participants, testosterone significantly improved the frequency of satisfying sexual events compared with placebo, along with desire, arousal, orgasm, pleasure, responsiveness and sexual self-image, and it reduced sexually related personal distress. That constellation — not one outcome but a coherent set of sexual-function outcomes — is why the finding held up to scrutiny and became the basis for a formal indication.

The same analysis looked for effects beyond sexual function and did not find them. Testosterone did not produce meaningful improvements in general wellbeing, mood or depressive symptoms, cognitive measures, bone mineral density, or body composition endpoints at the doses tested. This is a negative result from pooled randomized data, not an absence of research, which is a stronger statement than 'we don't know yet.' For those outcomes, the trials were run and testosterone did not beat placebo.

Route of delivery mattered for safety. The meta-analysis found that oral testosterone was associated with unfavorable changes in cholesterol — lower HDL and higher LDL and total cholesterol — while non-oral routes such as transdermal delivery were not. This is the specific evidence behind the guideline preference for creams, gels and patches over pills.

The trials also had limits that matter to anyone considering long-term use. They were mostly conducted in postmenopausal women, mostly short in duration, and collectively too small and too brief to detect rare or slow-developing outcomes such as breast cancer or cardiovascular events. Absence of harm across short trials is not evidence of safety over a decade.

How HSDD is actually diagnosed (it is not a blood test)

Hypoactive sexual desire disorder is a clinical diagnosis made by history. The core elements clinicians work from are: persistently absent or reduced sexual desire or sexual thoughts, present for a sustained period, that causes the woman herself personal distress, and that is not better explained by another cause. Every part of that definition does work. Low desire without personal distress is not a disorder. Low desire that is fully explained by a medication, an untreated depression, painful intercourse, or a relationship situation is not HSDD either.

The distress criterion is the one most often skipped in a rushed telehealth intake. A woman whose desire has declined but who is not troubled by it does not have a condition requiring treatment, and testosterone has nothing to offer her. Conversely, distress is the outcome the trials actually moved — the meta-analysis measured sexually related personal distress as an endpoint and found it fell.

The 'not better explained by' step is the substance of a good evaluation. A clinician should be asking about pain with penetration, vaginal dryness, sleep quality, current antidepressants (particularly SSRIs and SNRIs), thyroid symptoms, alcohol, chronic pain, mood, and the state of the relationship and the partner's sexual health. The International Society for the Study of Women's Sexual Health clinical practice guideline frames systemic testosterone as appropriate for postmenopausal HSDD after this kind of assessment, not instead of it.

There is no testosterone level that establishes the diagnosis. Blood androgen concentrations do not correlate reliably with sexual desire in women, and the Global Consensus Position Statement is explicit that testosterone levels should not be used to diagnose low desire or to select candidates for treatment. A clinic that runs a testosterone panel, calls the result 'low,' and prescribes on that basis has inverted the logic of the guideline.

What testosterone is not supported for

The gap between what testosterone is marketed for and what it has been shown to do is wide, and it is worth naming claim by claim. The Global Consensus Position Statement recommends against the use of testosterone in women for indications other than HSDD, including cognitive performance, cardiometabolic health, bone health, and general wellbeing, because the available evidence does not support them.

Two of these deserve specific attention because they are the most common sales pitches. Fatigue and low energy: pooled trial data did not show a benefit over placebo, and fatigue in midlife women has a long list of causes — disrupted sleep from night sweats, iron deficiency, thyroid disease, depression, sleep apnea — that are testable and often treatable. Bone density: testosterone is not a treatment for osteoporosis in women, and using it as one means the actual bone problem goes unmanaged while the years pass.

Hot flashes and night sweats are also outside its scope. The SWAN cohort followed vasomotor symptoms across the menopause transition and found they typically persist for years rather than months — this is a real, durable problem, but the therapy with evidence behind it is systemic estrogen, addressed in the 2022 hormone therapy position statement of The North American Menopause Society, not testosterone.

Similarly, vaginal dryness and pain with sex belong to genitourinary syndrome of menopause, which NAMS addressed in a dedicated 2020 position statement. Low-dose vaginal estrogen products, such as the estradiol vaginal ring whose FDA labeling covers moderate to severe urogenital symptoms due to postmenopausal atrophy, treat the tissue problem directly. Adding testosterone for desire while leaving pain untreated is a common sequencing error: pain suppresses desire, and no androgen fixes that.

Why there is no FDA-approved testosterone product for women — and what that means for you

The United States has no testosterone formulation approved by the FDA for use in women, for any indication. This is not because the compound is unstudied; it is a regulatory and commercial history. The consequence is immediate and practical: any testosterone a woman receives in the US arrives by one of two routes, and both introduce a problem that would not exist with an approved female product.

The first route is off-label prescribing of a product approved for men — typically a transdermal gel or cream — with instructions to use a small portion of the male dose. The second is a compounded preparation made by a pharmacy to a prescriber's specification. Compounded drugs are not FDA-approved, are not subject to the same manufacturing and potency verification requirements, and are not required to demonstrate batch-to-batch consistency in the way approved products are. The National Academies conducted a dedicated review of the clinical utility of compounded bioidentical hormone therapy, and ACOG issued a 2023 Clinical Consensus on compounded bioidentical menopausal hormone therapy — both examine the evidence and regulatory gaps in this space.

What this means for a patient is that the assurance you take for granted with an approved drug — that the dose on the label is the dose in the tube — is weaker here. It also means there is no FDA-reviewed label listing contraindications, interactions and warnings for female use, so the safety framing depends entirely on how well the individual prescriber follows society guidelines.

It also means insurance coverage is usually absent and the market skews toward cash-pay clinics. That commercial structure creates pressure to prescribe broadly and to frame testosterone as a general vitality treatment, which is precisely the framing the consensus statement rejects.

Off-label male products versus compounded preparations: the dosing hazard

The central hazard is arithmetic. Female physiologic testosterone dosing is a small fraction of a standard male daily dose. Male gels and creams are packaged in sachets, pumps and metered applicators calibrated to deliver a male dose in one or a few actuations. Splitting that down accurately by eye, at home, every day, is not a task those delivery systems were designed for, and small measuring errors translate into proportionally large dose errors.

Overshooting is not a theoretical concern. It is the mechanism behind the androgenic side effects that women actually report, and it is why guidelines put such emphasis on confirming blood levels stay within the normal premenopausal female range rather than simply asking how the patient feels.

Compounded creams solve the measuring problem by being formulated at a female-appropriate concentration, which is a genuine advantage, but they trade it for uncertainty about potency and consistency, which is the concern the National Academies and ACOG have documented. Neither option is clean. A reasonable prescriber will acknowledge this trade-off out loud rather than presenting compounding as a superior 'bioidentical' product.

Pellets and injections are a separate category and guidelines treat them separately. Both deliver a bolus that the patient cannot titrate or withdraw, and both are associated with blood concentrations above the premenopausal range. The Global Consensus Position Statement advises against formulations that result in supraphysiologic levels. With a cream, a side effect can be managed by reducing or stopping the dose tomorrow. With a pellet implanted under the skin, there is no dose-down and no off switch — you wait it out.

What monitoring should actually look like

Testosterone monitoring in women has one job, and it is not the one most people assume. It is not to confirm the level is 'optimal' or to chase a number in a target range. It is a safety check to confirm the dose has not pushed blood concentrations above the normal premenopausal female range. The Global Consensus Position Statement frames it exactly this way.

A defensible sequence looks like this: a baseline total testosterone before starting, primarily so there is a reference point and so that an unexpectedly high baseline can be investigated rather than treated; a repeat measurement after roughly three to six weeks of steady use to confirm the level has not overshot; and periodic rechecks thereafter, particularly after any dose change or if androgenic side effects appear.

A technical point worth knowing when you read your own results: standard immunoassays are unreliable at the low concentrations relevant to women, having been designed around male ranges. Liquid chromatography–tandem mass spectrometry is the reference method for accurate measurement at female levels. If your clinic is making dose decisions on an assay that cannot resolve the range in question, the number is decorative.

Free testosterone and sex hormone binding globulin are sometimes measured, but calculated free testosterone depends on the assumptions of the formula used, and none of these values diagnose the desire problem. Monitoring should also include a symptom review — acne, unwanted hair growth, hair thinning at the scalp, voice change, clitoral enlargement — because side effects sometimes appear before a level looks abnormal, and the patient's report outranks the lab in that case.

The time-limited trial: three to six months, then decide

Testosterone for HSDD is a trial, not a lifelong commitment made at the first appointment. Guidelines including the Global Consensus Position Statement describe reassessing the response at around three to six months and discontinuing if there has been no meaningful improvement in the desire problem. This is the single most important structural feature of appropriate prescribing, and the one most likely to be missing from a subscription-model clinic whose revenue depends on continuation.

The reason for a defined endpoint is that continuing an ineffective androgen accumulates side-effect risk with no offsetting benefit. Acne and hair changes are dose- and duration-related. Continuing indefinitely on the theory that it might eventually work is not how the therapy was studied.

'Meaningful improvement' should be defined before starting, in the patient's own terms — more spontaneous desire, more satisfying sexual events, less distress about the change. Vague improvement in energy or mood is not the target outcome and, given that the pooled trials did not show benefit on those measures, is not evidence the drug is working.

If the trial succeeds and treatment continues, it should continue at the lowest effective dose with periodic reassessment of whether it is still needed and still tolerated, not on autopilot. If the trial fails, the correct next step is to revisit the diagnosis rather than to raise the dose above the physiologic range — the response rate to escalation beyond that range is not established, and the androgenic risk is.

Side effects: what is dose-dependent and what may not reverse

The pooled randomized trials found the most consistent adverse effects to be acne and increased body or facial hair growth. These are dose-related, generally develop over time, and typically improve when the dose is reduced or stopped. They are the most common reasons women discontinue.

Scalp hair thinning in a female-pattern distribution is the side effect women tend to find most distressing, and it is the one that responds least quickly to stopping. If it develops, it is worth evaluating properly rather than assuming it is permanent — female pattern hair loss has established treatments. Topical minoxidil was shown effective for female pattern hair loss in a randomized placebo-controlled trial published in the Journal of the American Academy of Dermatology, and low-dose oral minoxidil has been reported in the dermatology literature as an option for female pattern hair loss, with a separate review examining its efficacy and safety profile for hair loss more generally.

Voice deepening and clitoral enlargement sit in a different category. They are associated with sustained supraphysiologic exposure rather than physiologic dosing, and they may not fully reverse after stopping. This asymmetry — most effects reverse, a few may not — is the entire practical argument for staying within the premenopausal range and for avoiding delivery methods that cannot be dialed back.

Lipid effects depend on route. The meta-analysis found unfavorable cholesterol changes with oral testosterone and not with non-oral delivery, which is why transdermal is the preferred route rather than a matter of convenience. Blood pressure, liver enzymes and existing cardiovascular risk factors are reasonable things to have on file before starting, particularly if a clinician is considering any oral formulation.

Before blaming testosterone: the differential that gets skipped

Low desire is the endpoint of many different problems, and the ones that respond to something other than testosterone are more common than the one that responds to testosterone. Running this differential is what separates a real evaluation from an intake form.

Pain. Genitourinary syndrome of menopause causes vaginal dryness, burning and pain with penetration, and desire predictably collapses in the presence of anticipated pain. NAMS addressed this in its 2020 position statement, and low-dose vaginal estrogen therapies — including the estradiol vaginal ring, whose FDA label covers moderate to severe urogenital symptoms due to postmenopausal atrophy — treat the tissue rather than the desire. Treating pain first sometimes resolves the desire complaint entirely.

Medications. SSRIs and SNRIs are a leading pharmacologic cause of reduced desire, delayed orgasm and blunted arousal. So can some other agents. This is a conversation about dose, timing or alternative agents with the prescribing clinician, not a reason to add a second drug on top.

Sleep and vasomotor symptoms. Persistent night sweats fragment sleep, and chronic sleep deprivation suppresses desire independently of any hormone. Given the SWAN finding that vasomotor symptoms typically last for years rather than months, this is not a self-limiting nuisance to wait out.

Mood, relationship and life context. Depression suppresses desire directly, and antidepressants can suppress it further — a loop worth naming explicitly. Relationship conflict, a partner's sexual dysfunction, caregiving load and untreated anxiety all belong in the same conversation. None of this makes low desire 'psychological' rather than 'real'; it makes the correct treatment something other than an androgen.

Reproductive stage. Where a woman actually is in the menopause transition matters for interpreting symptoms, and the Stages of Reproductive Aging Workshop +10 staging system is the standard framework clinicians use. Women with primary ovarian insufficiency, described by MedlinePlus, are a distinct group whose hormonal management differs from typical menopause and who should be managed accordingly.

Premenopausal women, and the limits of the evidence

The trials that produced the positive finding were conducted predominantly in postmenopausal women, and that is why the indication reads 'postmenopausal HSDD' rather than 'HSDD.' There is not equivalent randomized evidence in premenopausal women, and the Global Consensus Position Statement does not extend the indication to them.

This is not the same as saying testosterone cannot help a premenopausal woman. It means the question has not been answered by adequate trials, so anyone prescribing in that setting is doing so without the evidence base that supports postmenopausal use, and should say so plainly. Contraception and pregnancy avoidance become material considerations as well, since androgen exposure in pregnancy is not something to take casually.

There is also a separate group whose situation is often conflated: women who have had both ovaries surgically removed. Surgical menopause produces an abrupt change rather than a gradual transition, and the clinical picture differs from natural menopause. That difference is worth raising specifically with a clinician rather than assuming general menopause guidance transfers directly.

Finally, the honest limits on long-term safety apply to everyone. The randomized trials were short. They were not designed or sized to answer whether years of testosterone exposure affects breast cancer risk or cardiovascular outcomes in women. A clinician who tells you long-term safety is established is overstating what exists; a clinician who acknowledges the uncertainty and monitors accordingly is describing the situation correctly.

Pellets, and why guidelines single them out

Subcutaneous testosterone pellets are heavily marketed in the US as convenient — an insertion every few months instead of a daily cream. The convenience is real. So is the structural problem: once the pellet is in, the dose cannot be reduced, paused or reversed. If acne, hair loss or voice change develops in week three, there is nothing to adjust.

Pellets are also associated with blood testosterone concentrations above the normal premenopausal female range, which is precisely the exposure pattern that the potentially irreversible androgenic effects are linked to. The Global Consensus Position Statement advises against formulations that produce supraphysiologic levels, and the ISSWSH clinical practice guideline likewise favors transdermal delivery over implanted or injected forms.

Pellets are compounded products, so the National Academies' review of compounded bioidentical hormone therapy and ACOG's 2023 Clinical Consensus on compounded bioidentical menopausal hormone therapy both bear directly on them. The combination — unverified potency plus an undoable delivery method — is why they attract more guideline caution than any other route.

If a clinic leads with pellets, that is diagnostic information about the clinic. It is reasonable to ask directly why a route that guidelines advise against is being recommended, and what the plan would be if a side effect appeared a week after insertion.

How to evaluate a prescriber or telehealth clinic

The quality gradient in this market is steep, and a few questions separate the ends of it quickly. Ask what the diagnosis is. A clinician prescribing appropriately will name HSDD, describe the distress criterion, and be able to explain what other causes were ruled out. A clinician who says your levels are 'low for your age' and moves to prescribing has skipped the diagnosis.

Ask what happens at three to six months. The correct answer includes the possibility of stopping. If the model is an indefinite monthly subscription with no defined reassessment and no stopping criterion, the commercial design is working against the guideline.

Ask about route. Transdermal at physiologic dosing is what guidelines support. Oral is discouraged on lipid grounds. Pellets and injections are advised against. A prescriber who defaults to pellets should be able to explain that departure.

Ask what monitoring is planned and what assay is used, and ask what the plan is if hair thinning or voice change appears. And note what the clinic claims testosterone will do for you. If energy, mood, focus or body composition appear in the marketing, the clinic is selling outside the evidence, and that tells you how much weight to give everything else it says. This site does not sell or prescribe treatment; decisions about testosterone belong with a clinician who has taken a full history.

Questions to ask your clinician

Bring these to your appointment — they turn a vague visit into a decision.

• What exactly are you diagnosing me with, and what else did you rule out before landing on it?
• Are you prescribing a compounded product or an off-label male product, and how will I measure the dose accurately at home?
• What will my blood testosterone be checked with, when, and what number would make you lower or stop the dose?
• At what point do we decide this isn't working, and what specifically would count as improvement?
• What is your plan if I develop scalp hair thinning, voice change, or acne on this dose?
• Why are you recommending this route rather than a transdermal cream, given that guidelines advise against pellets and injections?
• Is my vaginal dryness or pain with sex being treated separately, since that can suppress desire on its own?

Frequently asked questions

Does testosterone help with fatigue or low energy in women?

No. The 2019 systematic review and meta-analysis in Lancet Diabetes & Endocrinology found no meaningful benefit of testosterone over placebo for general wellbeing or energy-related outcomes in women, and the Global Consensus Position Statement recommends against prescribing it for those indications. Persistent fatigue in midlife has causes that are testable — disrupted sleep from night sweats, thyroid disease, iron deficiency, depression, sleep apnea — and those deserve evaluation rather than an androgen prescription.

Is there an FDA-approved testosterone product for women in the United States?

No. No testosterone formulation is FDA-approved for women in the US for any indication. Prescriptions come either from off-label use of male products at reduced doses or from compounded preparations, which are not FDA-approved and are not held to the same potency-verification standards. ACOG issued a 2023 Clinical Consensus on compounded bioidentical menopausal hormone therapy and the National Academies conducted a dedicated review of compounded hormone therapy, both of which examine the regulatory gaps.

Should I get my testosterone level tested to see if I need treatment?

Blood testosterone levels do not diagnose low sexual desire in women, and the Global Consensus Position Statement is explicit that they should not be used to select candidates for treatment. Levels are measured for a different reason: to confirm a baseline before starting and to check that treatment has not pushed concentrations above the normal premenopausal range. If a clinic runs a panel, calls the number low, and prescribes on that basis, it has reversed the guideline logic.

Why are testosterone pellets discouraged for women?

Two reasons. Pellets are associated with blood testosterone concentrations above the normal premenopausal female range, and that supraphysiologic exposure is what the potentially irreversible effects — voice deepening, clitoral enlargement — are linked to. Second, once inserted, a pellet cannot be reduced, paused or removed, so a side effect appearing in week three has no remedy other than waiting. Guidelines advise against formulations producing supraphysiologic levels and favor transdermal delivery.

How long before I know whether testosterone is working?

Guidelines describe a defined trial with reassessment at roughly three to six months, and discontinuation if the desire problem has not meaningfully improved. Blood levels are typically rechecked earlier, around three to six weeks in, to confirm the dose has not overshot. Define what improvement means before you start — more spontaneous desire, more satisfying sexual events, less distress — because vague improvement in mood or energy is not the outcome the trials showed and is not evidence the drug is working.

Does testosterone help vaginal dryness or pain during sex?

That is a different problem with a different treatment. Vaginal dryness, burning and pain with penetration are features of genitourinary syndrome of menopause, addressed in the 2020 NAMS position statement, and low-dose vaginal estrogen products treat the tissue directly. Adding testosterone for desire while leaving pain untreated tends to fail, because anticipated pain suppresses desire regardless of hormone levels. Treat the pain first and reassess desire afterward.

Is testosterone safe for women long term?

Not established either way. The randomized trials that support use for postmenopausal HSDD were short and were not sized to detect rare or slow-developing outcomes such as breast cancer or cardiovascular events. The meta-analysis found no excess of serious adverse events, but that is a statement about limited short-term data, not about years of exposure. Oral formulations were associated with adverse cholesterol changes that non-oral routes were not, which is why transdermal delivery is preferred.

Can premenopausal women use testosterone for low desire?

The randomized evidence supporting testosterone was generated predominantly in postmenopausal women, which is why the Global Consensus Position Statement limits the indication to postmenopausal HSDD. Use in premenopausal women is not supported by equivalent trial evidence. That does not prove it cannot help; it means anyone prescribing in that setting is working outside the evidence base and should say so. Contraception also becomes a material consideration.

What are the most common side effects of testosterone in women?

Acne and increased facial or body hair are the most consistently reported effects in randomized trials, both dose- and duration-related, and both generally improve after reducing or stopping. Scalp hair thinning in a female-pattern distribution is slower to reverse and may warrant dermatologic treatment; topical minoxidil has randomized trial support for female pattern hair loss. Voice deepening and clitoral enlargement are linked to sustained supraphysiologic exposure and may not fully reverse.

Will testosterone protect my bones or improve muscle tone?

No. The pooled trial data did not show meaningful benefit for bone mineral density or body composition in women at the doses studied, and the Global Consensus Position Statement recommends against using testosterone for bone health. Using it as a bone treatment carries a specific hazard: the actual bone problem goes unmanaged while years pass. Bone density concerns should be evaluated and treated on their own terms.

Primary sources

1. Safety and efficacy of testosterone for women: a systematic review and meta-analysis. Lancet Diabetes Endocrinol, 2019. PMID 31353194.
2. Global Consensus Position Statement on the Use of Testosterone Therapy for Women. J Clin Endocrinol Metab, 2019. PMID 31498871.
3. International Society for the Study of Women's Sexual Health Clinical Practice Guideline. J Sex Med, 2021. PMID 33814355.
4. The 2022 hormone therapy position statement of The North American Menopause Society. Menopause, 2022. PMID 35797481.
5. The 2020 genitourinary syndrome of menopause position statement of NAMS. Menopause, 2020. PMID 32852449.
6. Compounded Bioidentical Menopausal Hormone Therapy: ACOG Clinical Consensus. Obstet Gynecol, 2023. PMID 37856860.
7. National Academies: The Clinical Utility of Compounded Bioidentical Hormone Therapy.
8. Duration of menopausal vasomotor symptoms over the menopause transition. JAMA Intern Med, 2015. PMID 25686030.
9. Executive summary of the Stages of Reproductive Aging Workshop +10. J Clin Endocrinol Metab, 2012. PMID 22344196.
10. A randomized, placebo-controlled trial of 5% and 2% topical minoxidil in female pattern hair loss. J Am Acad Dermatol, 2004. PMID 15034503.
11. Low-Dose Oral Minoxidil for Female Pattern Hair Loss. Skin Appendage Disord, 2020. PMID 32656239.
12. Oral minoxidil treatment for hair loss: a review of efficacy and safety. J Am Acad Dermatol, 2021. PMID 32622136.
13. FDA prescribing information (estradiol vaginal ring), 2019.
14. ACOG: Perimenopausal Bleeding and Bleeding After Menopause (patient FAQ).
15. MedlinePlus: Primary ovarian insufficiency.
16. Office on Women's Health: Menopause basics.