Feminising Hormone Therapy

Evidence Review · ER-005

1. Introduction

Feminising hormone therapy is the use of medication to shift a person’s hormonal profile from one dominated by testosterone to one dominated by oestrogen, in order to produce feminine physical characteristics. It is a common first medical step for trans women and for some non-binary people assigned male at birth who seek bodily change. Gender-affirming treatment is uncommon at the population level — one Dutch estimate is roughly one in 2,800 people assigned male at birth — although prevalence estimates vary by country and method and are rising.^[8] [Netherlands-specific estimate.]

Why hormone therapy is used

For many people, the mismatch between their gender and the sex characteristics their body developed at puberty is a source of sustained distress, described clinically as gender dysphoria (or, increasingly, gender incongruence). FHT aims to reduce that distress by changing the body so it better matches the person’s gender.

Goals of treatment

FHT works in two directions: it suppresses testosterone-driven features (spontaneous erections, oilier skin, ongoing male-pattern changes) and induces feminine features (breast tissue, a more feminine fat distribution, softer skin). The major guidelines — WPATH Standards of Care, eighth version (SOC-8, 2022) and the Endocrine Society clinical practice guideline (2017) — frame the treatment target as bringing oestradiol and testosterone into the ranges typical of premenopausal women.^[1,2]

Gender dysphoria and the decision to treat

SOC-8 recommends offering hormone therapy when gender incongruence is marked and sustained, moves away from requiring a formal psychiatric diagnosis as a gate, and recognises that trained non-specialists may prescribe, to widen access.^[1] This reflects a shift towards individualised, needs-based assessment.

Informed consent and shared decision-making

Because some effects of FHT are permanent (notably breast growth and reduced fertility), good practice centres on informed consent: understanding the expected effects, their timelines, the irreversible elements, the realistic limits, and the known risks before starting. Choosing the regimen and route together with a clinician, in light of goals and health profile, is the model endorsed by current guidelines.^[1,2] Fertility preservation is discussed before treatment (see Section 6).

2. How feminising hormone therapy works

Oestrogen: mechanism, receptors and effects

The active oestrogen used in FHT is 17β-oestradiol — the bio-identical hormone produced by the ovaries, and the form recommended by the Endocrine Society and WPATH.^[2] It acts mainly by binding oestrogen receptors (ERα and ERβ) inside target cells, where the hormone–receptor complex switches specific genes on or off. Tissues differ in how many receptors they carry, which is part of why oestrogen drives breast development and fat redistribution but does not alter the larynx once it has matured.

Oestradiol does two jobs. First, it directly stimulates feminine tissue changes. Second, by acting on the pituitary and hypothalamus it suppresses the gonadotrophins (LH and FSH) that drive the testes, which lowers the body’s own testosterone. Oestradiol can suppress testosterone substantially, and in some people sufficiently; many people, however, need an additional testosterone-lowering medication to reach the female range, depending on dose, route, adherence and the target used.

Testosterone suppression: why it matters

Many features people most want to change — oilier skin, spontaneous erections, ongoing male-pattern changes — are driven by testosterone and its more potent derivative dihydrotestosterone (DHT). Lowering testosterone is therefore central. Importantly, once testosterone is adequately suppressed there is no good evidence that pushing oestradiol to high levels adds feminisation or breast growth; in the largest prospective study, oestradiol level did not predict breast growth.^[3,18] Higher-than-needed oestradiol mainly adds risk, not benefit.

Different approaches to lowering testosterone

• Reduce production via pituitary suppression — a GnRH analogue, oestradiol’s own suppressive effect, or cyproterone acetate (also antigonadotrophic).
• Block the androgen receptor — spironolactone (a competitive androgen-receptor antagonist) and, partly, cyproterone (which combines central gonadotrophin suppression with peripheral receptor blockade).
• Block conversion to DHT — a 5α-reductase inhibitor (finasteride or dutasteride). This narrows testosterone’s action rather than lowering it, and its role in FHT is limited and poorly evidenced (Section 3).

After orchidectomy, the main testosterone source is gone, and anti-androgens can usually be stopped while oestradiol continues, often at a lower dose.^[2]

Time course: what changes first, what takes years

Feminisation is slow and follows a fairly consistent order, though magnitude varies. The earliest changes (1–3 months) are reduced libido and fewer spontaneous erections. Softer, less oily skin and the start of fat redistribution follow over 3–6 months. Breast growth begins within the first few months, is often greatest during the first year, and may continue more gradually thereafter; fat redistribution continues over the first one to two years. Muscle mass falls over the first year or two. Body hair becomes finer and slower over one to three years, but established facial hair is only modestly affected. The voice does not change, and the skeleton does not remodel. The detailed timeline is set out as a sidebar. An important caveat (Section 10): this widely reproduced timeline derives largely from expert consensus and clinical experience, not rigorous prospective measurement — breast development is the main outcome that has been measured carefully.^[2,3]

3. Medications

This section is descriptive, not a recommendation of any specific drug or dose.

Oral oestradiol

Evidence. Oral 17β-oestradiol (often as oestradiol valerate) effectively raises oestradiol levels; its feminising effects are well established.

Advantages. Inexpensive, convenient, easy to titrate.

Disadvantages. Oral oestrogen passes first through the liver, raising clotting-factor production more than non-oral routes; in cisgender postmenopausal women, oral oestrogen carries higher venous thromboembolism (VTE) risk than transdermal.^[19] [Best established in cisgender women.] This is separate from the very high VTE risk historically seen with ethinylestradiol, a synthetic oestrogen far more thrombogenic than 17β-oestradiol and no longer recommended for FHT.^[2,10]

Transdermal oestradiol (patch or gel)

Evidence. Achieves physiological oestradiol levels and avoids first-pass liver metabolism.

Advantages. Lower impact on liver-derived clotting factors; in cisgender postmenopausal women, transdermal oestrogen shows little or no increase in VTE risk, and it carries the lowest thrombogenic profile reported in trans women — though there are no head-to-head trials in trans women comparing it directly with oral oestradiol.^[19] Often preferred for people at higher clot risk, including older people and smokers.

Disadvantages. Skin irritation; patches can be visible or detach; full testosterone suppression on transdermal oestradiol alone can require relatively high doses.

Injectable oestradiol

UK context. Injectable oestradiol esters are used in some countries (notably the United States) but are not a standard NHS option and are far less common in UK practice. Injections can produce high peak oestradiol levels with peaks and troughs between doses; where used, they are effective, the main considerations being level fluctuation and repeat injections.^[2]

GnRH analogues (agonists)

GnRH analogues suppress pituitary gonadotrophin release and strongly reduce testicular testosterone. Widely used in the UK and parts of Europe, they are an effective, clean way to suppress testosterone, without the prolactin, liver and meningioma concerns of cyproterone. The main drawbacks are cost and the need for injection.^[1,2]

Spironolactone

Spironolactone is an aldosterone antagonist that also acts as a competitive androgen-receptor antagonist; on its own it does not reliably lower circulating testosterone, so its benefit in feminisation comes mainly from blocking androgen action at the tissue level. It is the most common anti-androgen in the United States. In a randomised comparison it was clearly less potent than cyproterone: at 12 weeks, about 19% of the spironolactone group reached the female testosterone range (<50 ng/dL) versus about 90% on cyproterone.^[15] Its main monitoring concerns are raised potassium (hyperkalaemia) and low blood pressure. A systematic review found that, while several anti-androgens lower testosterone more than spironolactone, no studies adequately compared them on the outcomes patients actually care about — breast development and hair — so the most feminising anti-androgen remains unknown; it also cautions that testosterone level is an imperfect surrogate, since spironolactone works largely by blocking the androgen receptor rather than lowering testosterone.^[28]

Cyproterone acetate

Cyproterone acetate (CPA) is a potent anti-androgen and antigonadotrophin used widely in Europe and the UK; it suppresses testosterone more effectively than spironolactone.^[15] Recognised harms include raised prolactin (dose-dependent and usually reversible — often asymptomatic, but occasionally causing galactorrhoea or headache and warranting a prolactin check), changes in liver enzymes with rare reports of hepatotoxicity, reduced HDL cholesterol, and — the most serious signal — an association between higher-dose and higher-cumulative-dose use and intracranial meningioma.^[12,16] The dose-response is steep: in a large French cohort, meningioma risk was not clearly raised at lower cumulative doses but rose sharply at higher cumulative exposure (hazard ratio ~11 at 36–60 g and ~22 at ≥60 g cumulative), with roughly a sevenfold increase overall.^[24] [Population flag: this cohort was predominantly cisgender women on high-dose cyproterone; a trans-specific systematic review documents the same association in trans women but cannot define absolute risk.^[25]] European regulators have advised using the lowest effective dose and considering alternatives where practicable; the clear direction of practice is towards minimising CPA dose.^[17]

Finasteride and dutasteride (5α-reductase inhibitors)

These block conversion of testosterone to the more potent DHT. They are sometimes added (for example, for scalp hair) but their role in FHT is limited and poorly evidenced: they do not lower testosterone itself, and once testosterone is well suppressed their added benefit is unclear. There is little high-quality trial evidence for their use specifically in trans women.

Progesterone

Progesterone is one of the most contested questions in FHT and has its own companion review. In brief: WPATH SOC-8 (2022) reviewed the evidence and was unable to find quality evidence that progestogens improve breast or areolar development.^[1] Since then, an Amsterdam randomised controlled trial — the first of its kind — has examined adding oral progesterone for breast development. Its design is published;^[13] its results have so far been reported at a 2025 conference (EPATH) and via the centre’s own institutional announcement, describing an increase in breast volume measured by 3D surface imaging (reported as an increase of up to about 30%, largest in those who also raised their oestradiol dose) alongside side effects including drowsiness, breast and nipple tenderness and mood changes.^[14] Despite the strength of these preliminary signals, the trial’s primary outcomes are not yet available as a peer-reviewed published paper — a current literature search returns only the protocol and the institutional announcement — and on that basis the centre involved is offering progesterone only within a research setting. The honest current position: progesterone’s benefit is plausible and now under proper investigation, but not yet established by peer-reviewed outcome data.

See companion review: Progesterone in Feminising Hormone Therapy.

Formulation and interaction cautions

• Avoid ethinylestradiol and conjugated equine oestrogens for FHT; use 17β-oestradiol, which is less thrombogenic and gives reliable measured levels.^[2,10]
• Oral versus sublingual oestradiol: taking oestradiol sublingually produces higher, shorter-lived peaks, which makes level-based monitoring less reliable; route should be agreed with a clinician.
• Biotin supplements (high-dose, not ordinary dietary amounts) can interfere with some hormone immunoassays and produce misleading results; stopping high-dose biotin before blood tests avoids this.
• Spironolactone interactions: caution with ACE inhibitors, angiotensin-receptor blockers, potassium supplements and other potassium-raising drugs, and in kidney impairment, because of hyperkalaemia risk.
• Cyproterone acetate: meningioma risk rises with higher dose and cumulative exposure; regulatory advice is to use the lowest effective dose and consider alternatives.^[12,17]

4. Expected physical changes

This is the section readers ask about most, and where realistic expectations matter most. The single most important message is that variability is the rule: timelines and especially the degree of change differ substantially, and for several outcomes no measurable factor reliably predicts the result.^[3]

Breast development

Expected change. Growth of glandular tissue and fat, producing an anatomically feminine breast that is, on average, smaller than the cisgender female average.

Timeline. Often begins around 3–6 months. In prospective data, the largest measured changes occur in the first year, most within the first six months; growth then slows but does not stop. Three-year prospective follow-up found that the breast–chest (tape-measure) difference plateaued after about nine months, while breast volume measured by 3D imaging continued to increase, more gradually, across the full three years.^[3,4] Older guidance and cross-sectional data cite continued, more gradual change up to around 2–3 years and a typical Tanner stage of 2–3.^[2,18]

Evidence quality. Moderate — the best-measured feminising outcome. In the largest prospective multicentre study, the mean breast–chest difference increased by about 7.9 cm after one year, and nearly half of participants (~49%) had less than an AAA cup at one year.^[3]

Variability. High and largely unexplained. Oestradiol level did not predict breast growth, and no clinical or laboratory parameter reliably did.^[3] One caveat bounds even this finding: the studies measured total serum oestradiol, not the bioavailable (free) fraction that SHBG shifts — so the variable most likely to matter biologically was not the one measured. Final size is modest on average; a substantial proportion later choose breast augmentation, which is a realistic part of the conversation, not a treatment failure.

Body-fat redistribution

Expected change. Fat shifts from the abdomen towards hips, buttocks and thighs; face and limbs soften.

Timeline. Onset ~3–6 months; maximum ~2–5 years.^[2]

Evidence quality. Low–Moderate — the direction is well supported: a systematic review of imaging studies found oestrogen consistently associated with increased fat mass and decreased lean mass in trans women, though the specific timeline figures come largely from expert-consensus tables.^[2,29]

Variability. High; influenced by genetics, baseline body composition and weight.

Muscle mass and strength

Expected change. Gradual decrease as testosterone falls.

Timeline. Onset ~3–6 months; maximum ~1–2 years.^[2]

Evidence quality. Low–Moderate. Variability. Affected by activity and resistance training, which partly offset the loss.

Skin, pores and oil production

Expected change. Skin typically becomes softer and less oily; sebum (oil) production falls as androgen activity decreases, and acne often improves; pores may look less prominent.

Timeline. Within the first few months (~3–6 months).^[2]

Evidence quality. Low–Moderate — direction is consistent and mechanistically clear (sebum is androgen-driven), but much support is self-report and clinical observation rather than objective measurement. [See companion review: Skin Effects of Oestrogen.]

Variability. Moderate; some people find skin becomes too dry.

Body hair and facial hair

Expected change. Body hair becomes finer, lighter and slower-growing. Facial hair is reduced much less and rarely disappears; permanent removal (electrolysis or laser) is usually needed for the beard.

Timeline. Body-hair changes over ~6–12 months, continuing beyond three years; facial hair only modestly affected at any point.^[2]

Evidence quality. Low–Moderate for body hair; the limited facial-hair effect is well recognised clinically. Variability. High; strongly influenced by genetic background.

Scalp hair

Expected change. Existing male-pattern loss often slows or stabilises; hormones do not reliably regrow lost hair.

Timeline. Stabilisation may begin within months; regrowth, where it occurs, is partial.^[2]

Evidence quality. Low–Moderate. Variability. High; depends on the stage of loss at the start.

Fertility

FHT suppresses sperm production; semen parameters typically become abnormal and azoospermia is common with sustained treatment. Recovery after stopping is variable and not guaranteed. This is covered as a counselling topic in Section 6.^[2]

Libido, erectile function and emotional effects

Reduced sexual desire and fewer spontaneous erections are among the earliest effects (1–3 months, maximum by 3–6 months); erectile function commonly decreases. These are partly reversible if treatment changes.^[2] Many people report improved mood, wellbeing and quality of life and reduced dysphoria; this evidence is mostly observational and uncontrolled, so it supports the direction of benefit more than its precise size. FHT should not be presented as a substitute for appropriate mental-health care where that is needed. [Evidence: Low–Moderate.]

5. What hormones cannot change

One of the most useful things to be clear about is what FHT does not do. Hormones change soft tissue (fat, skin, glandular breast) and physiology, but they do not remodel the adult skeleton or the voice. The features below are not altered by hormones and, if a person wishes to change them, require other interventions.

6. Before starting: assessment, suitability and fertility

The brief picture before treatment matters as much as monitoring during it. The specifics below reflect common guideline-based practice (chiefly the Endocrine Society 2017 guideline and standard hormone-prescribing practice); exact processes vary between services.^[2]

Baseline assessment

A pre-treatment review typically covers:

• History: general medical history; personal and family history of venous thromboembolism or thrombophilia; migraine (and specifically whether with aura); hypertension; smoking and other cardiovascular risk factors; liver disease; kidney disease; and current medications (for interactions).
• Fertility intentions and a discussion of preservation (below).
• Examination: baseline blood pressure and weight/BMI.
• Baseline blood tests: oestradiol and testosterone; and, guided by the planned regimen, prolactin, potassium and kidney function (where spironolactone is planned), and liver enzymes (where cyproterone is planned), alongside general health bloods as clinically indicated.^[2]

Contraindications and precautions

These are best understood as situations that require specialist discussion and individualised care — not automatic exclusion. The conditions below are commonly treated as precautions requiring individual risk assessment: active or recent venous thromboembolism; known thrombophilia; oestrogen-sensitive cancer; severe uncontrolled hypertension; severe liver disease; significant cardiovascular disease; migraine with aura; and the combination of smoking with older age. In several of these, choosing a transdermal route and the lowest effective dose, optimising the underlying risk factor, and involving the relevant specialist allow treatment to proceed safely in many people.^[2,19]

Important evidence note: much of the contraindication and precaution framework — particularly the migraine-with-aura/stroke signal — is extrapolated from data on combined hormonal contraception and menopausal hormone therapy in cisgender women, where the strongest stroke signal is tied to the older, more thrombogenic oestrogen ethinylestradiol rather than to 17β-oestradiol. These precautions are reasonable and widely applied, but the trans-specific evidence behind them is limited.

Fertility counselling

FHT suppresses sperm production, and recovery after stopping cannot be assumed. Sperm cryopreservation before starting treatment is therefore recommended and should be offered and discussed in advance, so that the option is not lost.^[2]

7. Monitoring

Monitoring confirms that treatment is working and catches problems early. The following reflects common guideline-based practice.^[2]

Hormone levels

The two core tests are oestradiol and testosterone. The usual aims are testosterone in the female range (commonly cited as below ~50 ng/dL, roughly <1.7 nmol/L) and oestradiol within the premenopausal female range (commonly ~100–200 pg/mL, about 367–734 pmol/L) rather than higher.^[2] The Endocrine Society suggests checking these every three months in the first year, then once or twice yearly when stable.^[2] Because oestradiol fluctuates and assays vary between laboratories, levels are interpreted alongside clinical progress rather than chased in isolation.^[18] This caution now has direct support: a 2025 systematic review of 49 studies found that the commonly cited 100–200 pg/mL oestradiol range lacks a primary evidence base and has not been shown to optimise feminisation or reduce adverse events — so it is best treated as a pragmatic guide, with adequate testosterone suppression the more meaningful aim.^[28,30] Part of the difficulty is technical: routine laboratories measure total oestradiol by immunoassay, which is variable and prone to cross-reactivity, and which does not capture changes in sex hormone-binding globulin (SHBG) that shift the biologically active free fraction — a further reason to weight clinical progress over any single target number.

Regimen-specific safety bloods

• On spironolactone: monitor potassium and kidney function (hyperkalaemia risk).
• On cyproterone acetate: monitor liver enzymes, and check prolactin at baseline and periodically (for example annually), since cyproterone raises prolactin dose-dependently; markedly or progressively elevated levels warrant further assessment, including for prolactinoma. Very high-frequency screening is not necessary, but baseline-and-periodic prolactin monitoring is a reasonable standard precaution.^[12,16]

Blood pressure, weight and bone

Blood pressure and weight are checked routinely. Adequate sex-hormone levels protect bone; the main bone risk is under-treatment (insufficient hormone levels, especially after gonadectomy). Trans women often have lower baseline bone density than cisgender comparators even before treatment; systematic-review data find no consistent bone loss on oestrogen therapy and, in places, a small increase in lumbar-spine density, though long-term fracture data remain limited.^[29] Bone-density scanning (DXA) is used selectively, guided by risk factors, rather than universally.^[2]

Cancer screening

Screening should follow the organs a person has and their hormone-exposure history, rather than a single rule, and should take account of age, duration of oestrogen exposure, family history and local guidelines. Breast-screening considerations apply to trans women who have developed breast tissue; the prostate is retained after vaginoplasty, so it remains relevant — though, as noted in Section 8, PSA interpretation is complicated by androgen suppression.^[5,9]

8. Risks

Two distinctions run through this section:

• Absolute versus relative risk. A “doubling” (relative) can still be a small change if the baseline (absolute) risk is low. Most risks below are relative increases on a low base.
• Old oestrogens versus modern oestradiol. Much alarming historical data comes from ethinylestradiol, far more thrombogenic than today’s 17β-oestradiol; applying those figures to modern regimens overstates risk.^[10,11]

The FHT evidence base is dominated by a few observational, mostly European cohorts, which cannot fully separate hormones from confounders such as smoking, HIV and socioeconomic factors.^[6,8]

Venous thromboembolism (VTE)

• The very high historical VTE rates (several per cent, and roughly twenty-fold above cisgender male controls in older data) were driven overwhelmingly by ethinylestradiol, now abandoned for FHT.^[10]
• With modern 17β-oestradiol the absolute risk is much lower. The ~1.6% figure is a cumulative incidence accrued over years of follow-up — in modern cohorts the average time from starting therapy to a first clot is measured in years, not months — and is not an annual risk; pooled estimates put the annual rate in the region of 2 per 1,000 person-years, albeit with wide uncertainty.^[22,31]
• Trans women on oestrogen still show a modestly higher VTE incidence than cisgender comparison groups in cohort studies.^[6,7]
• Route matters: transdermal oestrogen carries the lowest thrombogenic profile — firmly established in cisgender postmenopausal women, consistent with the more limited trans data, but not demonstrated head-to-head in trans women.^[19] [Population flag.]
• Risk concentrates around surgery and with additional risk factors (smoking, older age, prior clots, immobility, thrombophilia).
• During the peri-operative period, oestrogen management is individualised — clinicians may pause oestrogen before higher-risk procedures and use standard thromboprophylaxis, weighed against the symptom and dysphoria impact of stopping; practice varies and should be agreed with the surgical team.^[31]

Stroke and cardiovascular disease

A large Dutch cohort found that trans women on hormone therapy had a higher incidence of stroke and VTE than both reference men and reference women, and a higher incidence of myocardial infarction than reference women.^[6] A US cohort similarly found increased cardiovascular events.^[7] A 2024 meta-analysis (22 studies) puts numbers to this: in trans women versus cisgender men, stroke incidence was about 1.8% (relative risk 1.3) and VTE about 1.6% (relative risk 2.2), but myocardial infarction was about 1.2% with no increase (relative risk 1.0); overall cardiovascular risk was roughly 40% higher.^[22] The most recent Amsterdam analysis (2025) is consistent — VTE raised, but myocardial infarction not increased and, against general-population men, actually lower.^[23] These are relative increases on a low base, and the studies could not exclude confounding by smoking and lifestyle. Older long-term data linked ethinylestradiol specifically to a roughly three-fold increase in cardiovascular death — another reason it is no longer used.^[11]

Mortality

The longest follow-up, from the Amsterdam cohort (2,927 trans women over five decades), found overall mortality higher than expected compared with general-population men (standardised mortality ratio ~1.8) and women (~2.8), with cause-specific excess in cardiovascular disease, lung cancer, HIV-related disease and suicide. Critically, the authors concluded this pattern gives no indication of a specific harmful effect of hormone treatment itself, and instead points to the importance of monitoring and addressing comorbidities and lifestyle factors (such as smoking) and the social determinants of health.^[8]

Smoking and age

Smoking and increasing age most consistently amplify the cardiovascular and clotting risks of oestrogen. Smoking cessation is one of the highest-value interventions for anyone on FHT, and transdermal oestrogen is often preferred in smokers and older people.^[19]

Cancer

Breast. In a Dutch nationwide cohort of 2,260 trans women on hormones for a median of 18 years, breast cancer was about 46 times more common than in cisgender men but less common than in cisgender women (standardised incidence ratio 0.2; 15 invasive cases).^[5] The authors concluded that standard cisgender breast-screening guidance is appropriate. Screening in practice also depends on age, duration of oestrogen exposure, family history and local guidelines.

Prostate. The prostate is retained after genital surgery. In a cohort of 2,281 trans women (median follow-up 14 years), only six prostate cancers occurred, a lower incidence than in reference males (standardised incidence ratio 0.20) — consistent with the protective effect of low androgen levels.^[9] One screening caveat: androgen suppression substantially lowers PSA, so standard reference ranges are not validated under these conditions and a “normal” PSA may be falsely reassuring — results are better interpreted against the person’s own trend and clinical picture than against the usual cut-offs. Fixed numerical correction factors borrowed from cisgender men on 5α-reductase inhibitors have not been validated in trans women on feminising therapy, so a specific multiplier cannot be recommended.

Bone health

Adequate oestrogen protects bone; the principal risk is under-treatment, especially after gonadectomy if hormones are not continued adequately. Robust fracture-outcome data specific to trans women remain limited, so monitoring focuses on maintaining adequate hormone levels and using DXA selectively.^[2]

Liver

Modern 17β-oestradiol — especially transdermal — has minimal hepatotoxicity, unlike older synthetic oestrogens. The more relevant liver consideration is cyproterone acetate, which can raise liver enzymes and has rare reports of serious hepatotoxicity, supporting liver-enzyme monitoring on CPA.^[12]

Mental health

The best current synthesis is a 2023 systematic review of 46 studies, in which gender-affirming hormone therapy was consistently associated with reductions in depressive symptoms and psychological distress, while evidence for broad quality-of-life gains was inconsistent and the risk of bias across studies was highly variable.^[26] A prospective controlled study similarly found a significant reduction in gender dysphoria within six months of starting therapy, though quality-of-life domains did not shift significantly over that short window.^[27] So the direction of psychological benefit — especially for distress and dysphoria — is reasonably well supported, even if its precise size and durability are not, and FHT should not be presented as a substitute for appropriate mental-health care where that is needed.

This section discusses serious health risks, including suicide as a cause of death in cohort data. It is general information, not personal medical advice; the relevance of each risk differs greatly between individuals and should be discussed with your own clinician. (See the support note at the end of this article.)

9. Special situations

Older adults

Older people have a higher baseline cardiovascular and clotting risk, so the route and dose of oestrogen matter more. Transdermal oestradiol and the lowest effective dose are commonly preferred, cardiovascular risk factors are optimised, and treatment is individualised. As with menopausal hormone therapy in cisgender women, the presence of cardiovascular risk factors is generally a reason for care and route selection, not an automatic exclusion.^[19] [Largely extrapolated from cisgender HRT data.]

Non-binary people and partial feminisation

Not everyone wants full feminisation. SOC-8 explicitly recognises non-binary people, who may seek partial or specific changes.^[1] Regimens can be tailored to such goals — for example, lower doses, oestrogen without an anti-androgen, or an anti-androgen without oestrogen — accepting that the evidence base for partial or non-standard regimens is thinner than for standard full-dose feminisation, so expectations and monitoring should be discussed individually.

10. Evidence quality

A core aim of this review is to be explicit about how strong the evidence is for each claim. The ratings use a simple traffic-light scheme — 🟢 reasonably robust, 🟡 moderate, 🔴 limited — assigned by judging the type of evidence (randomised trial, prospective cohort, expert consensus), its consistency, how directly it applies to trans women, and the size of the effect relative to its uncertainty. These are editorial judgements, not formal GRADE ratings. For more on how we weigh evidence in this field — why randomised trials are scarce, why cohort studies dominate, and how we treat figures borrowed from cisgender populations — see Reading the Evidence.

11. Frequently asked questions

Can I choose patches over tablets? Often, yes — a shared decision with your clinician. Transdermal oestradiol avoids first-pass liver metabolism and carries the lowest clot-risk profile, so it is frequently preferred for people at higher clot risk, including smokers and older people.^[19]

Does progesterone improve breast growth? Not established. SOC-8 found no quality evidence of benefit; a first randomised trial has reported, at a conference and via a press release, increased breast volume, but the peer-reviewed outcome paper is not yet available, so this remains preliminary.^[1,13,14]

Can hormones change my facial structure? Hormones soften the face by redistributing fat but do not change the facial bones; bony change requires surgery (Section 5).

Can fertility return after stopping? Sometimes, but it cannot be relied upon — which is why sperm banking before starting is recommended.^[2]

How long before I see skin changes? Usually softer, less oily skin within the first few months (~3–6 months).^[2]

Why do results differ so much? Genuine biological variability. For breast development, the largest study found that no measurable factor — including oestradiol level — predicted the outcome.^[3]

Will higher oestrogen doses give more feminisation? Once testosterone is well suppressed, there is no good evidence that higher oestradiol adds feminisation or breast growth — it mainly adds risk.^[18]

12. Clinical bottom line

1. Oestradiol reliably produces feminising changes in most adults; the degree varies widely.^[2,3]
2. 17β-oestradiol is the appropriate oestrogen; ethinylestradiol is not used in modern FHT because of its higher clot risk.^[2,10]
3. Adequate testosterone suppression is central; very high oestradiol adds risk without clear added benefit once testosterone is suppressed.^[18]
4. Breast development is usually modest, begins within the first few months, is often greatest during the first year, may continue more gradually thereafter, and cannot be predicted from blood levels.^[3,4]
5. Route matters for clot risk: transdermal oestradiol is generally lowest-risk, though the trans-specific evidence is extrapolated.^[19]
6. Smoking substantially increases the cardiovascular and clotting risks of oestrogen; cessation is high-value.^[8,19]
7. Cardiovascular and clot risks are modestly raised in cohort studies on a low absolute base and are confounded; overall excess mortality shows no signal of a direct hormonal effect.^[6,8]
8. Breast cancer risk is higher than in cisgender men but lower than in cisgender women; prostate cancer is rare under androgen suppression; screening follows anatomy and exposure.^[5,9]
9. Cyproterone acetate is effective but best used at the lowest effective dose given prolactin, liver and meningioma signals.^[12,17]
10. Evidence for progesterone remains limited and is evolving; baseline assessment and regular monitoring improve the safety of any regimen.^[1,2,13,14]