Women's HRT: What It Is, How It Works, and What the Evidence Actually Says

What women's HRT is and who typically considers it

Women's HRT replaces the hormones, primarily estradiol and, when the uterus is present, progesterone, that the ovaries produce in declining and increasingly erratic amounts from perimenopause onward. Candidates typically include women with bothersome menopause symptoms, women with premature ovarian insufficiency (POI, ovarian failure before age 40), and women who have had surgical removal of both ovaries.

The direct answer: For women without contraindications who are within about 10 years of menopause or under roughly age 60, systemic HRT is the most effective treatment available for hot flashes, night sweats, and vaginal/urinary symptoms of menopause, and it also reduces bone loss. The tradeoff is a small absolute increase in breast cancer risk with combined estrogen-progestogen regimens, a risk that is smaller with estrogen-only therapy and appears to vary by progestogen type. Starting HRT for the first time well after age 60, or many years past menopause, changes this balance and requires more individualized risk assessment.

Perimenopause typically begins several years before the final menstrual period and involves fluctuating rather than simply declining estrogen, which is part of why vasomotor symptoms can be severe during this transition. Surgical menopause causes an abrupt hormone drop and tends to produce more severe symptoms than natural menopause; starting HRT soon after surgery, when appropriate, generally addresses this most directly.

POI is uncommon but clinically important because the resulting years of estrogen deficiency, if untreated, carry long-term cardiovascular and bone consequences beyond symptom burden. Professional guidance generally supports continuing HRT for women with POI at least until the average age of natural menopause, absent contraindications; the specific age threshold and duration should be confirmed with current specialty guidelines rather than treated as fixed here.

Contraindications commonly cited across guidelines include unexplained vaginal bleeding, active or recent estrogen-sensitive breast or endometrial cancer, active venous thromboembolism, and uncontrolled cardiovascular disease. Women with a personal history of cardiovascular disease, VTE, or a strong family or genetic (for example BRCA) cancer history need individualized risk assessment before starting, not a standard recommendation.

Estrogen-only versus combined therapy: why the uterus matters

The choice between estrogen-only and combined HRT is driven by whether the uterus is present. Women without a uterus can generally use estrogen alone. Women with a uterus need a progestogen added to protect the endometrium from unopposed estrogen stimulation, which otherwise raises the risk of endometrial hyperplasia and cancer.

Progestogen choice appears to matter for more than endometrial protection. Observational cohort data (notably French E3N-type cohort research) have suggested that combined regimens using micronized progesterone carry a more favorable breast cancer signal than regimens using older synthetic progestins such as medroxyprogesterone acetate. This is an important and influential finding in current prescribing patterns, but it comes from observational cohorts, not a randomized trial designed to compare progestogens head to head, and the exact effect size in the original literature should be verified before it is quoted as a precise number.

Route of estrogen delivery also matters biologically. Oral estradiol is metabolized through the liver on first pass and increases sex hormone-binding globulin and clotting factor production more than transdermal delivery does. Multiple observational studies have associated oral estrogen with higher venous thromboembolism risk than transdermal estrogen, and NICE guidance reflects this by favoring transdermal delivery for women with elevated VTE risk factors (NICE NG23, 2015, updated 2019). The exact magnitude of the oral-versus-transdermal VTE difference reported in specific studies could not be confirmed from the sources available for this draft and should be checked against the primary literature before being cited as a specific relative risk.

Vaginal estrogen is a separate category. Used at standard low doses for genitourinary syndrome of menopause (GSM), it produces minimal systemic absorption and is not considered systemic HRT requiring progestogen protection in most women. The FDA has approved multiple low-dose vaginal estrogen products for this indication (see the FDA drug approvals database for current product listings).

What the Women's Health Initiative did and did not show

The Women's Health Initiative (WHI) is the largest randomized controlled trial of combined HRT and remains central to how clinicians think about risk. It enrolled postmenopausal women, generally in their 50s to 70s, and compared conjugated equine estrogen plus a synthetic progestin against placebo. The trial was stopped early after an interim analysis showed increased breast cancer and coronary heart disease events in the combined-hormone group, which triggered a sharp and lasting drop in HRT prescribing worldwide.

Later age-stratified reanalysis of WHI data substantially changed the interpretation. When results were broken down by age at initiation and time since menopause, women who started hormone therapy earlier in the postmenopausal window showed a more favorable balance of outcomes, including improved all-cause mortality in some subgroups, than women who started later. This gave rise to the "timing hypothesis": estrogen's cardiovascular effect may depend on the state of the vasculature at the time therapy starts, protective in early, relatively healthy vessels, potentially destabilizing in vessels with established atherosclerosis.

A separate WHI ancillary study on cognition (the WHI Memory Study) found increased dementia risk in women aged 65 and older who started combined HRT, a population well outside the "timing window" described above. Smaller randomized trials in younger, recently menopausal women have generally not shown this signal over several years of follow-up. No randomized trial has demonstrated increased dementia risk from HRT started before age 60.

WHI's bone data are more consistent: the combined-hormone arm showed a meaningful reduction in hip and total fracture risk compared with placebo. This is one of the more reproducible findings across the HRT literature.

Because the exact hazard ratios, confidence intervals, and enrollment numbers attached to WHI and its reanalyses in the original source material could not be verified against a working citation for this draft, they have been described here in directional terms rather than as precise figures. An editor with database access should confirm and reinsert exact numbers with correct primary citations before publication.

Cardiovascular effects and the timing hypothesis

The timing hypothesis is the most useful single framework for reconciling why some studies show cardiovascular benefit from HRT and others show harm. In women who start therapy soon after menopause, when vascular disease is typically early or absent, observational and some trial data suggest estrogen may slow markers of atherosclerosis progression. In women who start a decade or more after menopause, when subclinical atherosclerosis is more likely already present, the same therapy has not shown benefit and, in the original WHI cohort (average age well into the 60s), was associated with early harm.

Professional cardiology statements have generally converged on two linked points: MHT started early in menopause has not been shown to increase cardiovascular risk and observational data suggest a possible reduction in coronary events in this group, but MHT is not an approved or recommended primary prevention strategy for cardiovascular disease at any age. Those two statements are compatible and both should guide expectations: HRT is prescribed for symptom relief and bone protection, not as heart disease prevention.

Breast cancer risk in plain terms

The most defensible summary from the trial and cohort literature is this: combined estrogen-progestogen HRT is associated with a small but real absolute increase in breast cancer risk that grows with duration of use, while estrogen-only HRT in women without a uterus has not shown a comparable increase and in the original WHI cohort trended toward fewer cancers over the trial period. This is an important asymmetry that is frequently lost in general reporting on "HRT and breast cancer."

Duration also matters within combined regimens. Multiple analyses in the literature describe an association between longer duration of combined HRT use and higher risk, with short-term use (roughly under a year to a few years) carrying a much smaller absolute risk than a decade or more of continuous use. The precise numeric risk added per year of use varies across studies and should be discussed with a clinician using current data rather than a fixed number quoted from any single source.

Framing the absolute risk in context is useful: several other common exposures, including alcohol intake and elevated body mass index, are associated with breast cancer risk increases of comparable or larger magnitude to combined HRT in some analyses. That comparison is meant to calibrate concern, not to dismiss the risk, and specific comparative figures should be verified in current literature before being used in patient counseling.

Genitourinary syndrome of menopause (GSM)

GSM, encompassing vaginal dryness, painful intercourse, urinary urgency, and recurrent urinary tract infections, is common after menopause and, unlike hot flashes, does not typically improve on its own over time.

Low-dose vaginal estrogen is first-line therapy for GSM in women who do not need or want systemic hormone therapy, and it can also be added for women already on systemic HRT whose genitourinary symptoms persist, since systemic therapy does not always deliver adequate local estrogen to vaginal tissue. Ospemifene, an oral selective estrogen receptor modulator, is an FDA-approved non-estrogen alternative for painful intercourse related to menopause. A body of randomized trial evidence, summarized in systematic reviews, supports local estrogen's efficacy for GSM symptoms with minimal systemic absorption at standard doses, though exact trial counts and effect sizes from the original source could not be verified here and should be confirmed against a current systematic review before citing specific numbers.

Bioidentical versus compounded hormones

"Bioidentical" describes a hormone molecule structurally identical to what the human body produces. FDA-approved bioidentical options exist and are widely used: 17-beta estradiol (available in patches, gels, and oral tablets under various brand names) and micronized progesterone (marketed as Prometrium and generics). These have standard manufacturing quality control and published efficacy and safety data reviewed by the FDA.

Compounded bioidentical HRT (cBHRT), prepared by compounding pharmacies, is a different category. It is not FDA-approved, is not subject to the same manufacturing oversight, and its potency and absorption can vary between batches and pharmacies. Endocrinology and menopause specialty organizations have generally advised against routine use of compounded hormone products when an FDA-approved equivalent exists, largely because of this lack of standardization and the absence of large controlled safety data specific to compounded formulations. Saliva or urine hormone testing, sometimes marketed to guide compounded HRT dosing, is not validated for that purpose; hormone levels measured this way vary with time of day and other factors and do not reliably reflect tissue-level hormone activity.

Women who want a molecularly "bioidentical" hormone do not need to use a compounded product to get one; FDA-approved estradiol and micronized progesterone already meet that definition.

A decision framework for thinking through HRT candidacy

The clinical literature does not reduce to a single yes/no answer for "should I take HRT." The following framework organizes the handful of factors that most change the benefit-risk calculation, based on the evidence patterns described above. It is an educational organizing tool, not a substitute for an individualized clinical evaluation, and it does not set a dose or make a diagnosis.

Step 1: Where are you relative to menopause?

• Within about 10 years of your final period, or under roughly age 60, and symptomatic: the benefit-risk ratio for systemic HRT is generally most favorable in the literature. This is the group most guideline statements are written for.
• More than about 10 years past menopause, or over 60, and starting HRT for the first time: benefit-risk is less clear, cardiovascular and clot risk are relatively higher, and the decision needs individualized specialist input rather than default initiation.

Step 2: Do you have a uterus?

• Yes: you need a progestogen alongside estrogen. Ask specifically whether micronized progesterone versus a synthetic progestin changes your personal risk profile, since the two are not interchangeable in the observational cancer-risk literature.
• No (prior hysterectomy): estrogen-only therapy is an option, with a different and generally more favorable breast cancer profile than combined therapy in trial data.

Step 3: Do you have clotting, cardiovascular, or hormone-sensitive cancer risk factors?

• History of VTE, stroke, known clotting disorder, or strong family/genetic breast or endometrial cancer risk: these are the factors that most often shift the route (transdermal over oral) or rule out systemic HRT altogether. This conversation should happen before a prescription, not after a side effect.
• No such history: route choice becomes more about tolerability and preference, though transdermal is still commonly preferred as a lower-risk default.

Step 4: What are you actually trying to treat?

• Vasomotor symptoms (hot flashes, night sweats) or bone protection: systemic HRT has the strongest evidence base.
• Vaginal dryness, painful intercourse, or urinary symptoms only: low-dose vaginal estrogen alone may be sufficient and avoids most systemic risk questions entirely.
• Mood, sleep, or cognitive concerns alone, without vasomotor symptoms: HRT's evidence for these outcomes is weaker and less consistent; a broader workup and non-hormonal options deserve equal consideration.

Step 5: What would make this an urgent, not routine, conversation?

• Unexplained vaginal bleeding, a new breast lump, leg swelling or pain suggestive of a clot, sudden chest pain, or neurological symptoms while on HRT are reasons to seek prompt medical evaluation rather than waiting for a scheduled follow-up.

Dosing patterns, monitoring, and stopping

Systemic estradiol is available in several delivery forms (patch, gel, spray, oral tablet) and clinicians typically start at a low dose and adjust based on symptom response, generally allowing several weeks at a given dose before judging effect, since hot flash improvement often appears within about a month while mood, sleep, and joint symptoms can take two to three months to improve fully. Specific starting and titration doses are prescriber decisions based on formulation, individual risk factors, and current product labeling; they are not provided here because dosing should always be individualized by a treating clinician rather than taken from a general reference article.

Progestogen is dosed in one of two patterns for women with a uterus: sequential (cyclical), giving progestogen for part of each month, generally used closer to the menopause transition when a woman may still have some uterine bleeding; or continuous combined, giving progestogen daily, generally used once a woman is further past her final period, with irregular spotting common in the first several months before bleeding typically settles.

Follow-up after starting HRT typically includes reassessment within the first few months to review symptom response and any abnormal bleeding. Bleeding outside the expected pattern on sequential therapy, or any bleeding at all on continuous combined therapy after the initial adjustment period, generally warrants evaluation, which may include pelvic ultrasound or endometrial sampling depending on findings, per a clinician's judgment.

There is no universal fixed stop date for HRT in appropriate candidates; guideline bodies generally favor an annual reassessment of the benefit-risk balance over an arbitrary time limit, particularly for women who remain symptomatic or who have POI. Stopping abruptly commonly leads to a return of vasomotor symptoms, sometimes more intense than before treatment; a gradual dose taper is commonly recommended in clinical practice, though no specific tapering schedule has strong randomized evidence behind it as superior to others.

Cognitive health: a genuinely unsettled area

Estrogen receptors are present throughout brain regions involved in memory, and animal research has demonstrated neuroprotective effects of estradiol at the cellular level. Whether this translates into a meaningful clinical cognitive benefit or protection against dementia in humans is not established.

Human cohort research has found that verbal memory can decline measurably during the menopause transition itself, and some research links this decline to hormone fluctuation during the transition rather than to absolute hormone level afterward. Separately, the WHI Memory Study found increased dementia risk with combined HRT started in women 65 and older, a population outside the "early initiation" window discussed above; smaller trials in younger, recently menopausal women have not replicated that harm signal over several years of follow-up, though this is reassurance about absence of harm in that group rather than proof of cognitive benefit.

Current major guideline statements do not support prescribing HRT specifically to prevent dementia in the general menopausal population, because randomized evidence for that specific purpose is insufficient. Observational data on estrogen-only therapy and Alzheimer's risk exist but are subject to healthy-user bias (women who continue long-term hormone therapy tend to differ systematically from those who do not) and should not be used, on their own, to justify starting HRT for brain protection. For women who are already appropriate candidates based on vasomotor or genitourinary symptoms, the cognitive evidence is broadly reassuring rather than a reason for concern, but it is not a distinct indication.

Non-hormonal alternatives

Not every woman is a candidate for systemic HRT, and some prefer to avoid it. No non-hormonal option matches HRT's effect size for moderate-to-severe vasomotor symptoms, but several have genuine trial support:

• Fezolinetant, a neurokinin-3 receptor antagonist, received FDA approval in 2023 for moderate-to-severe vasomotor symptoms of menopause and has randomized trial data supporting a meaningful reduction in hot flash frequency compared with placebo; it carries a labeled hepatotoxicity warning requiring liver function monitoring, which is worth confirming against current FDA labeling before starting.
• Low-dose paroxetine is the only FDA-approved antidepressant specifically for hot flashes, with trial data supporting a meaningful reduction in frequency. Venlafaxine, gabapentin, and clonidine have supportive off-label trial data for vasomotor symptoms as well.
• Cognitive behavioral therapy (CBT) programs designed for menopause have randomized trial evidence for reducing the distress and perceived severity of hot flashes, working by changing the experience of symptoms rather than eliminating the physiological event; for women with mild-to-moderate symptoms, or a preference to avoid medication, this effect has been described as clinically meaningful in trial reports.

What is established, what is plausible, and what is not established

Established by trial and guideline consensus: HRT is the most effective treatment for moderate-to-severe vasomotor symptoms and reduces fracture risk; combined HRT carries a small absolute increase in breast cancer risk that rises with duration of use; oral estrogen carries higher VTE risk than transdermal estrogen; low-dose vaginal estrogen is effective and largely non-systemic for GSM.

Plausible but not proven at the level of a definitive randomized trial: that starting HRT early confers a net cardiovascular benefit (supported by mechanistic and observational data, the "timing hypothesis," but not established as primary prevention); that micronized progesterone carries meaningfully lower breast cancer risk than synthetic progestins (supported mainly by observational cohorts); that early HRT initiation reduces long-term dementia risk in the general population.

Not established: that HRT should be prescribed specifically for cognitive protection or dementia prevention in women without other indications; that any single tapering schedule is superior for discontinuation; precise numeric risk figures for many of the associations above, several of which could not be verified against a working primary citation in the source material for this article and require confirmation from current guideline documents or the original trial publications before being used in patient-facing numeric claims.

References

1. National Institute for Health and Care Excellence. Menopause: diagnosis and management. NICE guideline NG23 (2015, updated 2019). https://www.nice.org.uk/guidance/ng23
2. U.S. Food and Drug Administration. Drug approvals and databases (use to verify current labeling for estrogen, progesterone, ospemifene, and fezolinetant products). https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm

Several claims in this article originate from widely cited studies, including the Women's Health Initiative and its later reanalyses, the KEEPS and ELITE trials, the SWAN cohort, and E3N cohort research on progestogen type and breast cancer risk. The specific PMID and DOI links associated with these studies in earlier versions of this article could not be verified as pointing to the correct papers, so exact figures (hazard ratios, confidence intervals, sample sizes) have been described in directional terms rather than restated as precise numbers. An editor with literature database access should locate and re-verify the primary publications before republishing this article with specific statistics attached to named citations.